weekly_nerf.md

Weekly Classified Neural Radiance Fields

Filter by classes:

all | dynamic | editing | fast | generalization | human | video | lighting | reconstruction | texture | semantic | pose-slam | others

Dec27 - Jan3, 2023

• Boosting UAV Tracking with Voxel-Based Trajectory-Aware Pre-Training, RAL2022 | [code]

  Siamese network-based object tracking has remarkably promoted the automatic capability for highly-maneuvered unmanned aerial vehicles (UAVs). However, the leading-edge tracking framework often depends on template matching, making it trapped when facing multiple views of object in consecutive frames. Moreover, the general image-level pretrained backbone can overfit to holistic representations, causing the misalignment to learn object-level properties in UAV tracking. To tackle these issues, this work presents TRTrack , a comprehensive framework to fully exploit the stereoscopic representation for UAV tracking. Specifically, a novel pre-training paradigm method is proposed. Through trajectory-aware reconstruction training (TRT), the capability of the backbone to extract stereoscopic structure feature is strengthened without any parameter increment. Accordingly, an innovative hierarchical self-attention Transformer is proposed to capture the local detail information and global structure knowledge. For optimizing the correlation map, we proposed a novel spatial correlation refinement (SCR) module, which promotes the capability of modeling the long-range spatial dependencies. Comprehensive experiments on three UAV challenging benchmarks demonstrate that the proposed TRTrack achieves superior UAV tracking performance in both precision and efficiency. Quantitative tests in real-world settings fully prove the effectiveness of our work.

Dec25 - Dec31, 2022

• Neural Radiance Fields from Sparse RGB-D Images for High-Quality View Synthesis, TPAMI2022 | [code]

  The recently proposed neural radiance fields (NeRF) use a continuous function formulated as a multi-layer perceptron (MLP) to model the appearance and geometry of a 3D scene. This enables realistic synthesis of novel views, even for scenes with view dependent appearance. Many follow-up works have since extended NeRFs in different ways. However, a fundamental restriction of the method remains that it requires a large number of images captured from densely placed viewpoints for high-quality synthesis and the quality of the results quickly degrades when the number of captured views is insufficient. To address this problem, we propose a novel NeRF-based framework capable of high-quality view synthesis using only a sparse set of RGB-D images, which can be easily captured using cameras and LiDAR sensors on current consumer devices. First, a geometric proxy of the scene is reconstructed from the captured RGB-D images. Renderings of the reconstructed scene along with precise camera parameters can then be used to pre-train a network. Finally, the network is fine-tuned with a small number of real captured images. We further introduce a patch discriminator to supervise the network under novel views during fine-tuning, as well as a 3D color prior to improve synthesis quality. We demonstrate that our method can generate arbitrary novel views of a 3D scene from as few as 6 RGB-D images. Extensive experiments show the improvements of our method compared with the existing NeRF-based methods, including approaches that also aim to reduce the number of input images.

Dec18 - Dec24, 2022

• Removing Objects From Neural Radiance Fields | [code]

  Neural Radiance Fields (NeRFs) are emerging as a ubiquitous scene representation that allows for novel view synthesis. Increasingly, NeRFs will be shareable with other people. Before sharing a NeRF, though, it might be desirable to remove personal information or unsightly objects. Such removal is not easily achieved with the current NeRF editing frameworks. We propose a framework to remove objects from a NeRF representation created from an RGB-D sequence. Our NeRF inpainting method leverages recent work in 2D image inpainting and is guided by a user-provided mask. Our algorithm is underpinned by a confidence based view selection procedure. It chooses which of the individual 2D inpainted images to use in the creation of the NeRF, so that the resulting inpainted NeRF is 3D consistent. We show that our method for NeRF editing is effective for synthesizing plausible inpaintings in a multi-view coherent manner. We validate our approach using a new and still-challenging dataset for the task of NeRF inpainting.

• iLabel: Revealing Objects in Neural Fields, RAL2022 | [code]

  A neural field trained with self-supervision to efficiently represent the geometry and colour of a 3D scene tends to automatically decompose it into coherent and accurate object-like regions, which can be revealed with sparse labelling interactions to produce a 3D semantic scene segmentation. Our real-time iLabel system takes input from a hand-held RGB-D camera, requires zero prior training data, and works in an ‘open set’ manner, with semantic classes defined on the fly by the user. iLabel's underlying model is a simple multilayer perceptron (MLP), trained from scratch to learn a neural representation of a single 3D scene. The model is updated continually and visualised in real-time, allowing the user to focus interactions to achieve extremely efficient semantic segmentation. A room-scale scene can be accurately labelled into 10+ semantic categories with around 100 clicks, taking less than 5 minutes. Quantitative labelling accuracy scales powerfully with the number of clicks, and rapidly surpasses standard pre-trained semantic segmentation methods. We also demonstrate a hierarchical labelling variant of iLabel and a ‘hands-free’ mode where the user only needs to supply label names for automatically-generated locations.

• Masked Wavelet Representation for Compact Neural Radiance Fields | [code]

  Neural radiance fields (NeRF) have demonstrated the potential of coordinate-based neural representation (neural fields or implicit neural representation) in neural rendering. However, using a multi-layer perceptron (MLP) to represent a 3D scene or object requires enormous computational resources and time. There have been recent studies on how to reduce these computational inefficiencies by using additional data structures, such as grids or trees. Despite the promising performance, the explicit data structure necessitates a substantial amount of memory. In this work, we present a method to reduce the size without compromising the advantages of having additional data structures. In detail, we propose using the wavelet transform on grid-based neural fields. Grid-based neural fields are for fast convergence, and the wavelet transform, whose efficiency has been demonstrated in high-performance standard codecs, is to improve the parameter efficiency of grids. Furthermore, in order to achieve a higher sparsity of grid coefficients while maintaining reconstruction quality, we present a novel trainable masking approach. Experimental results demonstrate that non-spatial grid coefficients, such as wavelet coefficients, are capable of attaining a higher level of sparsity than spatial grid coefficients, resulting in a more compact representation. With our proposed mask and compression pipeline, we achieved state-of-the-art performance within a memory budget of 2 MB. Our code is available at this https URL.

Dec11 - Dec17, 2022

• NeRF-Art: Text-Driven Neural Radiance Fields Stylization | [code]

  As a powerful representation of 3D scenes, the neural radiance field (NeRF) enables high-quality novel view synthesis from multi-view images. Stylizing NeRF, however, remains challenging, especially on simulating a text-guided style with both the appearance and the geometry altered simultaneously. In this paper, we present NeRF-Art, a text-guided NeRF stylization approach that manipulates the style of a pre-trained NeRF model with a simple text prompt. Unlike previous approaches that either lack sufficient geometry deformations and texture details or require meshes to guide the stylization, our method can shift a 3D scene to the target style characterized by desired geometry and appearance variations without any mesh guidance. This is achieved by introducing a novel global-local contrastive learning strategy, combined with the directional constraint to simultaneously control both the trajectory and the strength of the target style. Moreover, we adopt a weight regularization method to effectively suppress cloudy artifacts and geometry noises which arise easily when the density field is transformed during geometry stylization. Through extensive experiments on various styles, we demonstrate that our method is effective and robust regarding both single-view stylization quality and cross-view consistency. The code and more results can be found in our project page: this https URL.

Dec4 - Dec10, 2022

• 4K-NeRF: High Fidelity Neural Radiance Fields at Ultra High Resolutions | [code]

  In this paper, we present a novel and effective framework, named 4K-NeRF, to pursue high fidelity view synthesis on the challenging scenarios of ultra high resolutions, building on the methodology of neural radiance fields (NeRF). The rendering procedure of NeRF-based methods typically relies on a pixel wise manner in which rays (or pixels) are treated independently on both training and inference phases, limiting its representational ability on describing subtle details especially when lifting to a extremely high resolution. We address the issue by better exploring ray correlation for enhancing high-frequency details benefiting from the use of geometry-aware local context. Particularly, we use the view-consistent encoder to model geometric information effectively in a lower resolution space and recover fine details through the view-consistent decoder, conditioned on ray features and depths estimated by the encoder. Joint training with patch-based sampling further facilitates our method incorporating the supervision from perception oriented regularization beyond pixel wise loss. Quantitative and qualitative comparisons with modern NeRF methods demonstrate that our method can significantly boost rendering quality for retaining high-frequency details, achieving the state-of-the-art visual quality on 4K ultra-high-resolution scenario. Code Available at \url{this https URL}

• Diffusion Guided Domain Adaptation of Image Generators | [code]

  Can a text-to-image diffusion model be used as a training objective for adapting a GAN generator to another domain? In this paper, we show that the classifier-free guidance can be leveraged as a critic and enable generators to distill knowledge from large-scale text-to-image diffusion models. Generators can be efficiently shifted into new domains indicated by text prompts without access to groundtruth samples from target domains. We demonstrate the effectiveness and controllability of our method through extensive experiments. Although not trained to minimize CLIP loss, our model achieves equally high CLIP scores and significantly lower FID than prior work on short prompts, and outperforms the baseline qualitatively and quantitatively on long and complicated prompts. To our best knowledge, the proposed method is the first attempt at incorporating large-scale pre-trained diffusion models and distillation sampling for text-driven image generator domain adaptation and gives a quality previously beyond possible. Moreover, we extend our work to 3D-aware style-based generators and DreamBooth guidance.

• Ref-NPR: Reference-Based Non-Photorealistic Radiance Fields | [code]

  Existing 3D scene stylization methods employ an arbitrary style reference to transfer textures and colors as styles without establishing meaningful semantic correspondences. We present Reference-Based Non-Photorealistic Radiance Fields, i.e., Ref-NPR. It is a controllable scene stylization method utilizing radiance fields to stylize a 3D scene, with a single stylized 2D view taken as reference. To achieve decent results, we propose a ray registration process based on the stylized reference view to obtain pseudo-ray supervision in novel views, and exploit the semantic correspondence in content images to fill occluded regions with perceptually similar styles. Combining these operations, Ref-NPR generates non-photorealistic and continuous novel view sequences with a single reference while obtaining reasonable stylization in occluded regions. Experiments show that Ref-NPR significantly outperforms other scene and video stylization methods in terms of both visual quality and semantic correspondence. Code and data will be made publicly available.

• NeRDi: Single-View NeRF Synthesis with Language-Guided Diffusion as General Image Priors | [code]

  2D-to-3D reconstruction is an ill-posed problem, yet humans are good at solving this problem due to their prior knowledge of the 3D world developed over years. Driven by this observation, we propose NeRDi, a single-view NeRF synthesis framework with general image priors from 2D diffusion models. Formulating single-view reconstruction as an image-conditioned 3D generation problem, we optimize the NeRF representations by minimizing a diffusion loss on its arbitrary view renderings with a pretrained image diffusion model under the input-view constraint. We leverage off-the-shelf vision-language models and introduce a two-section language guidance as conditioning inputs to the diffusion model. This is essentially helpful for improving multiview content coherence as it narrows down the general image prior conditioned on the semantic and visual features of the single-view input image. Additionally, we introduce a geometric loss based on estimated depth maps to regularize the underlying 3D geometry of the NeRF. Experimental results on the DTU MVS dataset show that our method can synthesize novel views with higher quality even compared to existing methods trained on this dataset. We also demonstrate our generalizability in zero-shot NeRF synthesis for in-the-wild images.

• GARF:Geometry-Aware Generalized Neural Radiance Field | [code]

  Neural Radiance Field (NeRF) has revolutionized free viewpoint rendering tasks and achieved impressive results. However, the efficiency and accuracy problems hinder its wide applications. To address these issues, we propose Geometry-Aware Generalized Neural Radiance Field (GARF) with a geometry-aware dynamic sampling (GADS) strategy to perform real-time novel view rendering and unsupervised depth estimation on unseen scenes without per-scene optimization. Distinct from most existing generalized NeRFs, our framework infers the unseen scenes on both pixel-scale and geometry-scale with only a few input images. More specifically, our method learns common attributes of novel-view synthesis by an encoder-decoder structure and a point-level learnable multi-view feature fusion module which helps avoid occlusion. To preserve scene characteristics in the generalized model, we introduce an unsupervised depth estimation module to derive the coarse geometry, narrow down the ray sampling interval to proximity space of the estimated surface and sample in expectation maximum position, constituting Geometry-Aware Dynamic Sampling strategy (GADS). Moreover, we introduce a Multi-level Semantic Consistency loss (MSC) to assist more informative representation learning. Extensive experiments on indoor and outdoor datasets show that comparing with state-of-the-art generalized NeRF methods, GARF reduces samples by more than 25%, while improving rendering quality and 3D geometry estimation.

• Fast and Lightweight Scene Regressor for Camera Relocalization | [code]

  Camera relocalization involving a prior 3D reconstruction plays a crucial role in many mixed reality and robotics applications. Estimating the camera pose directly with respect to pre-built 3D models can be prohibitively expensive for several applications with limited storage and/or communication bandwidth. Although recent scene and absolute pose regression methods have become popular for efficient camera localization, most of them are computation-resource intensive and difficult to obtain a real-time inference with high accuracy constraints. This study proposes a simple scene regression method that requires only a multi-layer perceptron network for mapping scene coordinates to achieve accurate camera pose estimations. The proposed approach uses sparse descriptors to regress the scene coordinates, instead of a dense RGB image. The use of sparse features provides several advantages. First, the proposed regressor network is substantially smaller than those reported in previous studies. This makes our system highly efficient and scalable. Second, the pre-built 3D models provide the most reliable and robust 2D-3D matches. Therefore, learning from them can lead to an awareness of equivalent features and substantially improve the generalization performance. A detailed analysis of our approach and extensive evaluations using existing datasets are provided to support the proposed method. The implementation detail is available at this https URL

Nov27 - Dec3, 2022

• StegaNeRF: Embedding Invisible Information within Neural Radiance Fields | [code]

  Recent advances in neural rendering imply a future of widespread visual data distributions through sharing NeRF model weights. However, while common visual data (images and videos) have standard approaches to embed ownership or copyright information explicitly or subtly, the problem remains unexplored for the emerging NeRF format. We present StegaNeRF, a method for steganographic information embedding in NeRF renderings. We design an optimization framework allowing accurate hidden information extractions from images rendered by NeRF, while preserving its original visual quality. We perform experimental evaluations of our method under several potential deployment scenarios, and we further discuss the insights discovered through our analysis. StegaNeRF signifies an initial exploration into the novel problem of instilling customizable, imperceptible, and recoverable information to NeRF renderings, with minimal impact to rendered images. Project page: this https URL.

• QFF: Quantized Fourier Features for Neural Field Representations | [code]

  Multilayer perceptrons (MLPs) learn high frequencies slowly. Recent approaches encode features in spatial bins to improve speed of learning details, but at the cost of larger model size and loss of continuity. Instead, we propose to encode features in bins of Fourier features that are commonly used for positional encoding. We call these Quantized Fourier Features (QFF). As a naturally multiresolution and periodic representation, our experiments show that using QFF can result in smaller model size, faster training, and better quality outputs for several applications, including Neural Image Representations (NIR), Neural Radiance Field (NeRF) and Signed Distance Function (SDF) modeling. QFF are easy to code, fast to compute, and serve as a simple drop-in addition to many neural field representations.

• 3D-TOGO: Towards Text-Guided Cross-Category 3D Object Generation, AAAI2023 | [code]

  Text-guided 3D object generation aims to generate 3D objects described by user-defined captions, which paves a flexible way to visualize what we imagined. Although some works have been devoted to solving this challenging task, these works either utilize some explicit 3D representations (e.g., mesh), which lack texture and require post-processing for rendering photo-realistic views; or require individual time-consuming optimization for every single case. Here, we make the first attempt to achieve generic text-guided cross-category 3D object generation via a new 3D-TOGO model, which integrates a text-to-views generation module and a views-to-3D generation module. The text-to-views generation module is designed to generate different views of the target 3D object given an input caption. prior-guidance, caption-guidance and view contrastive learning are proposed for achieving better view-consistency and caption similarity. Meanwhile, a pixelNeRF model is adopted for the views-to-3D generation module to obtain the implicit 3D neural representation from the previously-generated views. Our 3D-TOGO model generates 3D objects in the form of the neural radiance field with good texture and requires no time-cost optimization for every single caption. Besides, 3D-TOGO can control the category, color and shape of generated 3D objects with the input caption. Extensive experiments on the largest 3D object dataset (i.e., ABO) are conducted to verify that 3D-TOGO can better generate high-quality 3D objects according to the input captions across 98 different categories, in terms of PSNR, SSIM, LPIPS and CLIP-score, compared with text-NeRF and Dreamfields.

• LatentSwap3D: Semantic Edits on 3D Image GANs | [code]

  Recent 3D-aware GANs rely on volumetric rendering techniques to disentangle the pose and appearance of objects, de facto generating entire 3D volumes rather than single-view 2D images from a latent code. Complex image editing tasks can be performed in standard 2D-based GANs (e.g., StyleGAN models) as manipulation of latent dimensions. However, to the best of our knowledge, similar properties have only been partially explored for 3D-aware GAN models. This work aims to fill this gap by showing the limitations of existing methods and proposing LatentSwap3D, a model-agnostic approach designed to enable attribute editing in the latent space of pre-trained 3D-aware GANs. We first identify the most relevant dimensions in the latent space of the model controlling the targeted attribute by relying on the feature importance ranking of a random forest classifier. Then, to apply the transformation, we swap the top-K most relevant latent dimensions of the image being edited with an image exhibiting the desired attribute. Despite its simplicity, LatentSwap3D provides remarkable semantic edits in a disentangled manner and outperforms alternative approaches both qualitatively and quantitatively. We demonstrate our semantic edit approach on various 3D-aware generative models such as pi-GAN, GIRAFFE, StyleSDF, MVCGAN, EG3D and VolumeGAN, and on diverse datasets, such as FFHQ, AFHQ, Cats, MetFaces, and CompCars. The project page can be found: \url{this https URL}.

• DiffRF: Rendering-Guided 3D Radiance Field Diffusion | [code]

  We introduce DiffRF, a novel approach for 3D radiance field synthesis based on denoising diffusion probabilistic models. While existing diffusion-based methods operate on images, latent codes, or point cloud data, we are the first to directly generate volumetric radiance fields. To this end, we propose a 3D denoising model which directly operates on an explicit voxel grid representation. However, as radiance fields generated from a set of posed images can be ambiguous and contain artifacts, obtaining ground truth radiance field samples is non-trivial. We address this challenge by pairing the denoising formulation with a rendering loss, enabling our model to learn a deviated prior that favours good image quality instead of trying to replicate fitting errors like floating artifacts. In contrast to 2D-diffusion models, our model learns multi-view consistent priors, enabling free-view synthesis and accurate shape generation. Compared to 3D GANs, our diffusion-based approach naturally enables conditional generation such as masked completion or single-view 3D synthesis at inference time.

• NeuWigs: A Neural Dynamic Model for Volumetric Hair Capture and Animation | [code]

  The capture and animation of human hair are two of the major challenges in the creation of realistic avatars for the virtual reality. Both problems are highly challenging, because hair has complex geometry and appearance, as well as exhibits challenging motion. In this paper, we present a two-stage approach that models hair independently from the head to address these challenges in a data-driven manner. The first stage, state compression, learns a low-dimensional latent space of 3D hair states containing motion and appearance, via a novel autoencoder-as-a-tracker strategy. To better disentangle the hair and head in appearance learning, we employ multi-view hair segmentation masks in combination with a differentiable volumetric renderer. The second stage learns a novel hair dynamics model that performs temporal hair transfer based on the discovered latent codes. To enforce higher stability while driving our dynamics model, we employ the 3D point-cloud autoencoder from the compression stage for de-noising of the hair state. Our model outperforms the state of the art in novel view synthesis and is capable of creating novel hair animations without having to rely on hair observations as a driving signal. Project page is here this https URL.

• SparseFusion: Distilling View-conditioned Diffusion for 3D Reconstruction | [code]

  We propose SparseFusion, a sparse view 3D reconstruction approach that unifies recent advances in neural rendering and probabilistic image generation. Existing approaches typically build on neural rendering with re-projected features but fail to generate unseen regions or handle uncertainty under large viewpoint changes. Alternate methods treat this as a (probabilistic) 2D synthesis task, and while they can generate plausible 2D images, they do not infer a consistent underlying 3D. However, we find that this trade-off between 3D consistency and probabilistic image generation does not need to exist. In fact, we show that geometric consistency and generative inference can be complementary in a mode-seeking behavior. By distilling a 3D consistent scene representation from a view-conditioned latent diffusion model, we are able to recover a plausible 3D representation whose renderings are both accurate and realistic. We evaluate our approach across 51 categories in the CO3D dataset and show that it outperforms existing methods, in both distortion and perception metrics, for sparse-view novel view synthesis.

• Mixed Neural Voxels for Fast Multi-view Video Synthesis | [code]

  Synthesizing high-fidelity videos from real-world multi-view input is challenging because of the complexities of real-world environments and highly dynamic motions. Previous works based on neural radiance fields have demonstrated high-quality reconstructions of dynamic scenes. However, training such models on real-world scenes is time-consuming, usually taking days or weeks. In this paper, we present a novel method named MixVoxels to better represent the dynamic scenes with fast training speed and competitive rendering qualities. The proposed MixVoxels represents the 4D dynamic scenes as a mixture of static and dynamic voxels and processes them with different networks. In this way, the computation of the required modalities for static voxels can be processed by a lightweight model, which essentially reduces the amount of computation, especially for many daily dynamic scenes dominated by the static background. To separate the two kinds of voxels, we propose a novel variation field to estimate the temporal variance of each voxel. For the dynamic voxels, we design an inner-product time query method to efficiently query multiple time steps, which is essential to recover the high-dynamic motions. As a result, with 15 minutes of training for dynamic scenes with inputs of 300-frame videos, MixVoxels achieves better PSNR than previous methods. Codes and trained models are available at this https URL

• Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generation | [code]

  A diffusion model learns to predict a vector field of gradients. We propose to apply chain rule on the learned gradients, and back-propagate the score of a diffusion model through the Jacobian of a differentiable renderer, which we instantiate to be a voxel radiance field. This setup aggregates 2D scores at multiple camera viewpoints into a 3D score, and repurposes a pretrained 2D model for 3D data generation. We identify a technical challenge of distribution mismatch that arises in this application, and propose a novel estimation mechanism to resolve it. We run our algorithm on several off-the-shelf diffusion image generative models, including the recently released Stable Diffusion trained on the large-scale LAION dataset.

• Neural Subspaces for Light Fields, TVCG2022 | [code]

  We introduce a framework for compactly representing light field content with the novel concept of neural subspaces. While the recently proposed neural light field representation achieves great compression results by encoding a light field into a single neural network, the unified design is not optimized for the composite structures exhibited in light fields. Moreover, encoding every part of the light field into one network is not ideal for applications that require rapid transmission and decoding. We recognize this problem's connection to subspace learning. We present a method that uses several small neural networks, specializing in learning the neural subspace for a particular light field segment. Moreover, we propose an adaptive weight sharing strategy among those small networks, improving parameter efficiency. In effect, this strategy enables a concerted way to track the similarity among nearby neural subspaces by leveraging the layered structure of neural networks. Furthermore, we develop a soft-classification technique to enhance the color prediction accuracy of neural representations. Our experimental results show that our method better reconstructs the light field than previous methods on various light field scenes. We further demonstrate its successful deployment on encoding light fields with irregular viewpoint layout and dynamic scene content.

• 3D-LDM: Neural Implicit 3D Shape Generation with Latent Diffusion Models | [code]

  Diffusion models have shown great promise for image generation, beating GANs in terms of generation diversity, with comparable image quality. However, their application to 3D shapes has been limited to point or voxel representations that can in practice not accurately represent a 3D surface. We propose a diffusion model for neural implicit representations of 3D shapes that operates in the latent space of an auto-decoder. This allows us to generate diverse and high quality 3D surfaces. We additionally show that we can condition our model on images or text to enable image-to-3D generation and text-to-3D generation using CLIP embeddings. Furthermore, adding noise to the latent codes of existing shapes allows us to explore shape variations.

• SinGRAF: Learning a 3D Generative Radiance Field for a Single Scene | [code]

  Generative models have shown great promise in synthesizing photorealistic 3D objects, but they require large amounts of training data. We introduce SinGRAF, a 3D-aware generative model that is trained with a few input images of a single scene. Once trained, SinGRAF generates different realizations of this 3D scene that preserve the appearance of the input while varying scene layout. For this purpose, we build on recent progress in 3D GAN architectures and introduce a novel progressive-scale patch discrimination approach during training. With several experiments, we demonstrate that the results produced by SinGRAF outperform the closest related works in both quality and diversity by a large margin.

• NeAF: Learning Neural Angle Fields for Point Normal Estimation, AAAI2023 | [code]

  Normal estimation for unstructured point clouds is an important task in 3D computer vision. Current methods achieve encouraging results by mapping local patches to normal vectors or learning local surface fitting using neural networks. However, these methods are not generalized well to unseen scenarios and are sensitive to parameter settings. To resolve these issues, we propose an implicit function to learn an angle field around the normal of each point in the spherical coordinate system, which is dubbed as Neural Angle Fields (NeAF). Instead of directly predicting the normal of an input point, we predict the angle offset between the ground truth normal and a randomly sampled query normal. This strategy pushes the network to observe more diverse samples, which leads to higher prediction accuracy in a more robust manner. To predict normals from the learned angle fields at inference time, we randomly sample query vectors in a unit spherical space and take the vectors with minimal angle values as the predicted normals. To further leverage the prior learned by NeAF, we propose to refine the predicted normal vectors by minimizing the angle offsets. The experimental results with synthetic data and real scans show significant improvements over the state-of-the-art under widely used benchmarks.

• SNAF: Sparse-view CBCT Reconstruction with Neural Attenuation Fields | [code]

  Cone beam computed tomography (CBCT) has been widely used in clinical practice, especially in dental clinics, while the radiation dose of X-rays when capturing has been a long concern in CBCT imaging. Several research works have been proposed to reconstruct high-quality CBCT images from sparse-view 2D projections, but the current state-of-the-arts suffer from artifacts and the lack of fine details. In this paper, we propose SNAF for sparse-view CBCT reconstruction by learning the neural attenuation fields, where we have invented a novel view augmentation strategy to overcome the challenges introduced by insufficient data from sparse input views. Our approach achieves superior performance in terms of high reconstruction quality (30+ PSNR) with only 20 input views (25 times fewer than clinical collections), which outperforms the state-of-the-arts. We have further conducted comprehensive experiments and ablation analysis to validate the effectiveness of our approach.

• NeRFInvertor: High Fidelity NeRF-GAN Inversion for Single-shot Real Image Animation | [code]

  Nerf-based Generative models have shown impressive capacity in generating high-quality images with consistent 3D geometry. Despite successful synthesis of fake identity images randomly sampled from latent space, adopting these models for generating face images of real subjects is still a challenging task due to its so-called inversion issue. In this paper, we propose a universal method to surgically fine-tune these NeRF-GAN models in order to achieve high-fidelity animation of real subjects only by a single image. Given the optimized latent code for an out-of-domain real image, we employ 2D loss functions on the rendered image to reduce the identity gap. Furthermore, our method leverages explicit and implicit 3D regularizations using the in-domain neighborhood samples around the optimized latent code to remove geometrical and visual artifacts. Our experiments confirm the effectiveness of our method in realistic, high-fidelity, and 3D consistent animation of real faces on multiple NeRF-GAN models across different datasets.

• Differentiable Rendering Using RGBXY Derivatives and Optimal Transport, ToG2022 | [code]

  Traditional differentiable rendering approaches are usually hard to converge in inverse rendering optimizations, especially when initial and target object locations are not so close. Inspired by Lagrangian fluid simulation, we present a novel differentiable rendering method to address this problem. We associate each screen-space pixel with the visible 3D geometric point covered by the center of the pixel and compute derivatives on geometric points rather than on pixels. We refer to the associated geometric points as point proxies of pixels. For each point proxy, we compute its 5D RGBXY derivatives which measures how its 3D RGB color and 2D projected screen-space position change with respect to scene parameters. Furthermore, in order to capture global and long-range object motions, we utilize optimal transport based pixel matching to design a more sophisticated loss function. We have conducted experiments to evaluate the effectiveness of our proposed method on various inverse rendering applications and have demonstrated superior convergence behavior compared to state-of-the-art baselines.

• Efficient Light Probes for Real-Time Global Illumination, SIGGRAPH-Asia2022 | [code]

  Reproducing physically-based global illumination (GI) effects has been a long-standing demand for many real-time graphical applications. In pursuit of this goal, many recent engines resort to some form of light probes baked in a precomputation stage. Unfortunately, the GI effects stemming from the precomputed probes are rather limited due to the constraints in the probe storage, representation or query. In this paper, we propose a new method for probe-based GI rendering which can generate a wide range of GI effects, including glossy reflection with multiple bounces, in complex scenes. The key contributions behind our work include a gradient-based search algorithm and a neural image reconstruction method. The search algorithm is designed to reproject the probes' contents to any query viewpoint, without introducing parallax errors, and converges fast to the optimal solution. The neural image reconstruction method, based on a dedicated neural network and several G-buffers, tries to recover high-quality images from low-quality inputs due to limited resolution or (potential) low sampling rate of the probes. This neural method makes the generation of light probes efficient. Moreover, a temporal reprojection strategy and a temporal loss are employed to improve temporal stability for animation sequences. The whole pipeline runs in realtime (>30 frames per second) even for high-resolution (1920×1080) outputs, thanks to the fast convergence rate of the gradient-based search algorithm and a light-weight design of the neural network. Extensive experiments on multiple complex scenes have been conducted to show the superiority of our method over the state-of-the-arts.

• LaplacianFusion: Detailed 3D Clothed-Human Body Reconstruction, SIGGRAPH-Asia2022 | [code]

  We propose LaplacianFusion, a novel approach that reconstructs detailed and controllable 3D clothed-human body shapes from an input depth or 3D point cloud sequence. The key idea of our approach is to use Laplacian coordinates, well-known differential coordinates that have been used for mesh editing, for representing the local structures contained in the input scans, instead of implicit 3D functions or vertex displacements used previously. Our approach reconstructs a controllable base mesh using SMPL, and learns a surface function that predicts Laplacian coordinates representing surface details on the base mesh. For a given pose, we first build and subdivide a base mesh, which is a deformed SMPL template, and then estimate Laplacian coordinates for the mesh vertices using the surface function. The final reconstruction for the pose is obtained by integrating the estimated Laplacian coordinates as a whole. Experimental results show that our approach based on Laplacian coordinates successfully reconstructs more visually pleasing shape details than previous methods. The approach also enables various surface detail manipulations, such as detail transfer and enhancement.

• QuadStream: A Quad-Based Scene Streaming Architecture for Novel Viewpoint Reconstruction, ToG2022 | [code]

  Streaming rendered 3D content over a network to a thin client device, such as a phone or a VR/AR headset, brings high-fidelity graphics to platforms where it would not normally possible due to thermal, power, or cost constraints. Streamed 3D content must be transmitted with a representation that is both robust to latency and potential network dropouts. Transmitting a video stream and reprojecting to correct for changing viewpoints fails in the presence of disocclusion events; streaming scene geometry and performing high-quality rendering on the client is not possible on limited-power mobile GPUs. To balance the competing goals of disocclusion robustness and minimal client workload, we introduce QuadStream, a new streaming content representation that reduces motion-to-photon latency by allowing clients to efficiently render novel views without artifacts caused by disocclusion events. Motivated by traditional macroblock approaches to video codec design, we decompose the scene seen from positions in a view cell into a series of quad proxies, or view-aligned quads from multiple views. By operating on a rasterized G-Buffer, our approach is independent of the representation used for the scene itself; the resulting QuadStream is an approximate geometric representation of the scene that can be reconstructed by a thin client to render both the current view and nearby adjacent views. Our technical contributions are an efficient parallel quad generation, merging, and packing strategy for proxy views covering potential client movement in a scene; a packing and encoding strategy that allows masked quads with depth information to be transmitted as a frame-coherent stream; and an efficient rendering approach for rendering our QuadStream representation into entirely novel views on thin clients. We show that our approach achieves superior quality compared both to video data streaming methods, and to geometry-based streaming.

• DINER: Depth-aware Image-based NEural Radiance Fields | [code]

  We present Depth-aware Image-based NEural Radiance fields (DINER). Given a sparse set of RGB input views, we predict depth and feature maps to guide the reconstruction of a volumetric scene representation that allows us to render 3D objects under novel views. Specifically, we propose novel techniques to incorporate depth information into feature fusion and efficient scene sampling. In comparison to the previous state of the art, DINER achieves higher synthesis quality and can process input views with greater disparity. This allows us to capture scenes more completely without changing capturing hardware requirements and ultimately enables larger viewpoint changes during novel view synthesis. We evaluate our method by synthesizing novel views, both for human heads and for general objects, and observe significantly improved qualitative results and increased perceptual metrics compared to the previous state of the art. The code will be made publicly available for research purposes.

• Reconstructing Hand-Held Objects from Monocular Video, SIGGRAPH-Asia2022 | [code]

  This paper presents an approach that reconstructs a hand-held object from a monocular video. In contrast to many recent methods that directly predict object geometry by a trained network, the proposed approach does not require any learned prior about the object and is able to recover more accurate and detailed object geometry. The key idea is that the hand motion naturally provides multiple views of the object and the motion can be reliably estimated by a hand pose tracker. Then, the object geometry can be recovered by solving a multi-view reconstruction problem. We devise an implicit neural representation-based method to solve the reconstruction problem and address the issues of imprecise hand pose estimation, relative hand-object motion, and insufficient geometry optimization for small objects. We also provide a newly collected dataset with 3D ground truth to validate the proposed approach. The dataset and code will be released at https://dihuangdh.github.io/hhor.

• Dr.3D: Adapting 3D GANs to Artistic Drawings, SIGGRAPH-Asia2022 | [code]

  While 3D GANs have recently demonstrated the high-quality synthesis of multi-view consistent images and 3D shapes, they are mainly restricted to photo-realistic human portraits. This paper aims to extend 3D GANs to a different, but meaningful visual form: artistic portrait drawings. However, extending existing 3D GANs to drawings is challenging due to the inevitable geometric ambiguity present in drawings. To tackle this, we present Dr.3D, a novel adaptation approach that adapts an existing 3D GAN to artistic drawings. Dr.3D is equipped with three novel components to handle the geometric ambiguity: a deformation-aware 3D synthesis network, an alternating adaptation of pose estimation and image synthesis, and geometric priors. Experiments show that our approach can successfully adapt 3D GANs to drawings and enable multi-view consistent semantic editing of drawings.

• Efficient Neural Radiance Fields for Interactive Free-viewpoint Video, SIGGRAPH-Asia2022 | [code]

  This paper aims to tackle the challenge of efficiently producing interactive free-viewpoint videos. Some recent works equip neural radiance fields with image encoders, enabling them to generalize across scenes. When processing dynamic scenes, they can simply treat each video frame as an individual scene and perform novel view synthesis to generate free-viewpoint videos. However, their rendering process is slow and cannot support interactive applications. A major factor is that they sample lots of points in empty space when inferring radiance fields. We propose a novel scene representation, called ENeRF, for the fast creation of interactive free-viewpoint videos. Specifically, given multi-view images at one frame, we first build the cascade cost volume to predict the coarse geometry of the scene. The coarse geometry allows us to sample few points near the scene surface, thereby significantly improving the rendering speed. This process is fully differentiable, enabling us to jointly learn the depth prediction and radiance field networks from RGB images. Experiments on multiple benchmarks show that our approach exhibits competitive performance while being at least 60 times faster than previous generalizable radiance field methods.

• NeuLighting: Neural Lighting for Free Viewpoint Outdoor Scene Relighting with Unconstrained Photo Collections, SIGGRAPH-Asia2022 | [code]

  We propose NeuLighting, a new framework for free viewpoint outdoor scene relighting from a sparse set of unconstrained in-the-wild photo collections. Our framework represents all the scene components as continuous functions parameterized by MLPs that take a 3D location and the lighting condition as input and output reflectance and necessary outdoor illumination properties. Unlike object-level relighting methods which often leverage training images with controllable and consistent indoor illumination, we concentrate on the more challenging outdoor situation where all the images are captured under arbitrary unknown illumination. The key to our method includes a neural lighting representation that compresses the per-image illumination into a disentangled latent vector, and a new free viewpoint relighting scheme that is robust to arbitrary lighting variations across images. The lighting representation is compressive to explain a wide range of illumination and can be easily fed into the query-based NeuLighting framework, enabling efficient shading effect evaluation under any kind of novel illumination. Furthermore, to produce high-quality cast shadows, we estimate the sun visibility map to indicate the shadow regions according to the scene geometry and the sun direction. Thanks to the flexible and explainable neural lighting representation, our system supports outdoor relighting with many different illumination sources, including natural images, environment maps, and time-lapse videos. The high-fidelity renderings under novel views and illumination prove the superiority of our method against state-of-the-art relighting solutions.

• Lightweight Neural Basis Functions for All-Frequency Shading, SIGGRAPH-Asia2022 | [code]

  Basis functions provide both the abilities for compact representation and the properties for efficient computation. Therefore, they are pervasively used in rendering to perform all-frequency shading. However, common basis functions, including spherical harmonics (SH), wavelets, and spherical Gaussians (SG) all have their own limitations, such as low-frequency for SH, not rotationally invariant for wavelets, and no multiple product support for SG. In this paper, we present neural basis functions, an implicit and data-driven set of basis functions that circumvents the limitations with all desired properties. We first introduce a representation neural network that takes any general 2D spherical function (e.g. environment lighting, BRDF, and visibility) as input and projects it onto the latent space as coefficients of our neural basis functions. Then, we design several lightweight neural networks that perform different types of computation, giving our basis functions different computational properties such as double/triple product integrals and rotations. We demonstrate the practicality of our neural basis functions by integrating them into all-frequency shading applications, showing that our method not only achieves a compression rate of and 10 × -40 × better performance than wavelets at equal quality, but also renders all-frequency lighting effects in real-time without the aforementioned limitations from classic basis functions.

• DeepMVSHair: Deep Hair Modeling from Sparse Views, SIGGRAPH-Asia2022 | [code]

  We present DeepMVSHair, the first deep learning-based method for multi-view hair strand reconstruction. The key component of our pipeline is HairMVSNet, a differentiable neural architecture which represents a spatial hair structure as a continuous 3D hair growing direction field implicitly. Specifically, given a 3D query point, we decide its occupancy value and direction from observed 2D structure features. With the query point’s pixel-aligned features from each input view, we utilize a view-aware transformer encoder to aggregate anisotropic structure features to an integrated representation, which is decoded to yield 3D occupancy and direction at the query point. HairMVSNet effectively gathers multi-view hair structure features and preserves high-frequency details based on this implicit representation. Guided by HairMVSNet, our hair-growing algorithm produces results faithful to input multi-view images. We propose a novel image-guided multi-view strand deformation algorithm to enrich modeling details further. Extensive experiments show that the results by our sparse-view method are comparable to those by state-of-the-art dense multi-view methods and significantly better than those by single-view and sparse-view methods. In addition, our method is an order of magnitude faster than previous multi-view hair modeling methods.

• A Light Touch Approach to Teaching Transformers Multi-view Geometry | [code]

  Transformers are powerful visual learners, in large part due to their conspicuous lack of manually-specified priors. This flexibility can be problematic in tasks that involve multiple-view geometry, due to the near-infinite possible variations in 3D shapes and viewpoints (requiring flexibility), and the precise nature of projective geometry (obeying rigid laws). To resolve this conundrum, we propose a "light touch" approach, guiding visual Transformers to learn multiple-view geometry but allowing them to break free when needed. We achieve this by using epipolar lines to guide the Transformer's cross-attention maps, penalizing attention values outside the epipolar lines and encouraging higher attention along these lines since they contain geometrically plausible matches. Unlike previous methods, our proposal does not require any camera pose information at test-time. We focus on pose-invariant object instance retrieval, where standard Transformer networks struggle, due to the large differences in viewpoint between query and retrieved images. Experimentally, our method outperforms state-of-the-art approaches at object retrieval, without needing pose information at test-time.

• Fast-SNARF: A Fast Deformer for Articulated Neural Fields | [code]

  Neural fields have revolutionized the area of 3D reconstruction and novel view synthesis of rigid scenes. A key challenge in making such methods applicable to articulated objects, such as the human body, is to model the deformation of 3D locations between the rest pose (a canonical space) and the deformed space. We propose a new articulation module for neural fields, Fast-SNARF, which finds accurate correspondences between canonical space and posed space via iterative root finding. Fast-SNARF is a drop-in replacement in functionality to our previous work, SNARF, while significantly improving its computational efficiency. We contribute several algorithmic and implementation improvements over SNARF, yielding a speed-up of 150×. These improvements include voxel-based correspondence search, pre-computing the linear blend skinning function, and an efficient software implementation with CUDA kernels. Fast-SNARF enables efficient and simultaneous optimization of shape and skinning weights given deformed observations without correspondences (e.g. 3D meshes). Because learning of deformation maps is a crucial component in many 3D human avatar methods and since Fast-SNARF provides a computationally efficient solution, we believe that this work represents a significant step towards the practical creation of 3D virtual humans.

• Non-uniform Sampling Strategies for NeRF on 360° images , BMVC2022 | [code]

  In recent years, the performance of novel view synthesis using perspective images has dramatically improved with the advent of neural radiance fields (NeRF). This study proposes two novel techniques that effectively build NeRF for 360{\textdegree} omnidirectional images. Due to the characteristics of a 360{\textdegree} image of ERP format that has spatial distortion in their high latitude regions and a 360{\textdegree} wide viewing angle, NeRF's general ray sampling strategy is ineffective. Hence, the view synthesis accuracy of NeRF is limited and learning is not efficient. We propose two non-uniform ray sampling schemes for NeRF to suit 360{\textdegree} images - distortion-aware ray sampling and content-aware ray sampling. We created an evaluation dataset Synth360 using Replica and SceneCity models of indoor and outdoor scenes, respectively. In experiments, we show that our proposal successfully builds 360{\textdegree} image NeRF in terms of both accuracy and efficiency. The proposal is widely applicable to advanced variants of NeRF. DietNeRF, AugNeRF, and NeRF++ combined with the proposed techniques further improve the performance. Moreover, we show that our proposed method enhances the quality of real-world scenes in 360{\textdegree} images. Synth360: this https URL.

• High-fidelity 3D GAN Inversion by Pseudo-multi-view Optimization | [code]

  We present a high-fidelity 3D generative adversarial network (GAN) inversion framework that can synthesize photo-realistic novel views while preserving specific details of the input image. High-fidelity 3D GAN inversion is inherently challenging due to the geometry-texture trade-off in high-fidelity 3D inversion, where overfitting to a single view input image often damages the estimated geometry during the latent optimization. To solve this challenge, we propose a novel pipeline that builds on the pseudo-multi-view estimation with visibility analysis. We keep the original textures for the visible parts and utilize generative priors for the occluded parts. Extensive experiments show that our approach achieves advantageous reconstruction and novel view synthesis quality over state-of-the-art methods, even for images with out-of-distribution textures. The proposed pipeline also enables image attribute editing with the inverted latent code and 3D-aware texture modification. Our approach enables high-fidelity 3D rendering from a single image, which is promising for various applications of AI-generated 3D content.

Nov20 - Nov26, 2022

• ResNeRF: Geometry-Guided Residual Neural Radiance Field for Indoor Scene Novel View Synthesis | [code]

  We represent the ResNeRF, a novel geometry-guided two-stage framework for indoor scene novel view synthesis. Be aware of that a good geometry would greatly boost the performance of novel view synthesis, and to avoid the geometry ambiguity issue, we propose to characterize the density distribution of the scene based on a base density estimated from scene geometry and a residual density parameterized by the geometry. In the first stage, we focus on geometry reconstruction based on SDF representation, which would lead to a good geometry surface of the scene and also a sharp density. In the second stage, the residual density is learned based on the SDF learned in the first stage for encoding more details about the appearance. In this way, our method can better learn the density distribution with the geometry prior for high-fidelity novel view synthesis while preserving the 3D structures. Experiments on large-scale indoor scenes with many less-observed and textureless areas show that with the good 3D surface, our method achieves state-of-the-art performance for novel view synthesis.

• RUST: Latent Neural Scene Representations from Unposed Imagery | [code]

  Inferring the structure of 3D scenes from 2D observations is a fundamental challenge in computer vision. Recently popularized approaches based on neural scene representations have achieved tremendous impact and have been applied across a variety of applications. One of the major remaining challenges in this space is training a single model which can provide latent representations which effectively generalize beyond a single scene. Scene Representation Transformer (SRT) has shown promise in this direction, but scaling it to a larger set of diverse scenes is challenging and necessitates accurately posed ground truth data. To address this problem, we propose RUST (Really Unposed Scene representation Transformer), a pose-free approach to novel view synthesis trained on RGB images alone. Our main insight is that one can train a Pose Encoder that peeks at the target image and learns a latent pose embedding which is used by the decoder for view synthesis. We perform an empirical investigation into the learned latent pose structure and show that it allows meaningful test-time camera transformations and accurate explicit pose readouts. Perhaps surprisingly, RUST achieves similar quality as methods which have access to perfect camera pose, thereby unlocking the potential for large-scale training of amortized neural scene representations.

• Unsupervised Continual Semantic Adaptation through Neural Rendering | [code]

  An increasing amount of applications rely on data-driven models that are deployed for perception tasks across a sequence of scenes. Due to the mismatch between training and deployment data, adapting the model on the new scenes is often crucial to obtain good performance. In this work, we study continual multi-scene adaptation for the task of semantic segmentation, assuming that no ground-truth labels are available during deployment and that performance on the previous scenes should be maintained. We propose training a Semantic-NeRF network for each scene by fusing the predictions of a segmentation model and then using the view-consistent rendered semantic labels as pseudo-labels to adapt the model. Through joint training with the segmentation model, the Semantic-NeRF model effectively enables 2D-3D knowledge transfer. Furthermore, due to its compact size, it can be stored in a long-term memory and subsequently used to render data from arbitrary viewpoints to reduce forgetting. We evaluate our approach on ScanNet, where we outperform both a voxel-based baseline and a state-of-the-art unsupervised domain adaptation method.

• ShadowNeuS: Neural SDF Reconstruction by Shadow Ray Supervision | [code]

  By supervising camera rays between a scene and multi-view image planes, NeRF reconstructs a neural scene representation for the task of novel view synthesis. On the other hand, shadow rays between the light source and the scene have yet to be considered. Therefore, we propose a novel shadow ray supervision scheme that optimizes both the samples along the ray and the ray location. By supervising shadow rays, we successfully reconstruct a neural SDF of the scene from single-view pure shadow or RGB images under multiple lighting conditions. Given single-view binary shadows, we train a neural network to reconstruct a complete scene not limited by the camera's line of sight. By further modeling the correlation between the image colors and the shadow rays, our technique can also be effectively extended to RGB inputs. We compare our method with previous works on challenging tasks of shape reconstruction from single-view binary shadow or RGB images and observe significant improvements. The code and data will be released.

• Dynamic Neural Portraits, WACV2023 | [code]

  We present Dynamic Neural Portraits, a novel approach to the problem of full-head reenactment. Our method generates photo-realistic video portraits by explicitly controlling head pose, facial expressions and eye gaze. Our proposed architecture is different from existing methods that rely on GAN-based image-to-image translation networks for transforming renderings of 3D faces into photo-realistic images. Instead, we build our system upon a 2D coordinate-based MLP with controllable dynamics. Our intuition to adopt a 2D-based representation, as opposed to recent 3D NeRF-like systems, stems from the fact that video portraits are captured by monocular stationary cameras, therefore, only a single viewpoint of the scene is available. Primarily, we condition our generative model on expression blendshapes, nonetheless, we show that our system can be successfully driven by audio features as well. Our experiments demonstrate that the proposed method is 270 times faster than recent NeRF-based reenactment methods, with our networks achieving speeds of 24 fps for resolutions up to 1024 x 1024, while outperforming prior works in terms of visual quality.

• ScanNeRF: a Scalable Benchmark for Neural Radiance Fields, WACV2023 | [code]

  In this paper, we propose the first-ever real benchmark thought for evaluating Neural Radiance Fields (NeRFs) and, in general, Neural Rendering (NR) frameworks. We design and implement an effective pipeline for scanning real objects in quantity and effortlessly. Our scan station is built with less than 500$ hardware budget and can collect roughly 4000 images of a scanned object in just 5 minutes. Such a platform is used to build ScanNeRF, a dataset characterized by several train/val/test splits aimed at benchmarking the performance of modern NeRF methods under different conditions. Accordingly, we evaluate three cutting-edge NeRF variants on it to highlight their strengths and weaknesses. The dataset is available on our project page, together with an online benchmark to foster the development of better and better NeRFs.

• DiffusionSDF: Conditional Generative Modeling of Signed Distance Functions | [code]

  Probabilistic diffusion models have achieved state-of-the-art results for image synthesis, inpainting, and text-to-image tasks. However, they are still in the early stages of generating complex 3D shapes. This work proposes DiffusionSDF, a generative model for shape completion, single-view reconstruction, and reconstruction of real-scanned point clouds. We use neural signed distance functions (SDFs) as our 3D representation to parameterize the geometry of various signals (e.g., point clouds, 2D images) through neural networks. Neural SDFs are implicit functions and diffusing them amounts to learning the reversal of their neural network weights, which we solve using a custom modulation module. Extensive experiments show that our method is capable of both realistic unconditional generation and conditional generation from partial inputs. This work expands the domain of diffusion models from learning 2D, explicit representations, to 3D, implicit representations.

• Immersive Neural Graphics Primitives | [code]

  Neural radiance field (NeRF), in particular its extension by instant neural graphics primitives, is a novel rendering method for view synthesis that uses real-world images to build photo-realistic immersive virtual scenes. Despite its potential, research on the combination of NeRF and virtual reality (VR) remains sparse. Currently, there is no integration into typical VR systems available, and the performance and suitability of NeRF implementations for VR have not been evaluated, for instance, for different scene complexities or screen resolutions. In this paper, we present and evaluate a NeRF-based framework that is capable of rendering scenes in immersive VR allowing users to freely move their heads to explore complex real-world scenes. We evaluate our framework by benchmarking three different NeRF scenes concerning their rendering performance at different scene complexities and resolutions. Utilizing super-resolution, our approach can yield a frame rate of 30 frames per second with a resolution of 1280x720 pixels per eye. We discuss potential applications of our framework and provide an open source implementation online.

• BAD-NeRF: Bundle Adjusted Deblur Neural Radiance Fields | [code]

  Neural Radiance Fields (NeRF) have received considerable attention recently, due to its impressive capability in photo-realistic 3D reconstruction and novel view synthesis, given a set of posed camera images. Earlier work usually assumes the input images are in good quality. However, image degradation (e.g. image motion blur in low-light conditions) can easily happen in real-world scenarios, which would further affect the rendering quality of NeRF. In this paper, we present a novel bundle adjusted deblur Neural Radiance Fields (BAD-NeRF), which can be robust to severe motion blurred images and inaccurate camera poses. Our approach models the physical image formation process of a motion blurred image, and jointly learns the parameters of NeRF and recovers the camera motion trajectories during exposure time. In experiments, we show that by directly modeling the real physical image formation process, BAD-NeRF achieves superior performance over prior works on both synthetic and real datasets.

• Peekaboo: Text to Image Diffusion Models are Zero-Shot Segmentors | [code]

  Recent diffusion-based generative models combined with vision-language models are capable of creating realistic images from natural language prompts. While these models are trained on large internet-scale datasets, such pre-trained models are not directly introduced to any semantic localization or grounding. Most current approaches for localization or grounding rely on human-annotated localization information in the form of bounding boxes or segmentation masks. The exceptions are a few unsupervised methods that utilize architectures or loss functions geared towards localization, but they need to be trained separately. In this work, we explore how off-the-shelf diffusion models, trained with no exposure to such localization information, are capable of grounding various semantic phrases with no segmentation-specific re-training. An inference time optimization process is introduced, that is capable of generating segmentation masks conditioned on natural language. We evaluate our proposal Peekaboo for unsupervised semantic segmentation on the Pascal VOC dataset. In addition, we evaluate for referring segmentation on the RefCOCO dataset. In summary, we present a first zero-shot, open-vocabulary, unsupervised (no localization information), semantic grounding technique leveraging diffusion-based generative models with no re-training. Our code will be released publicly.

• OReX: Object Reconstruction from Planner Cross-sections Using Neural Fields | [code]

  Reconstructing 3D shapes from planar cross-sections is a challenge inspired by downstream applications like medical imaging and geographic informatics. The input is an in/out indicator function fully defined on a sparse collection of planes in space, and the output is an interpolation of the indicator function to the entire volume. Previous works addressing this sparse and ill-posed problem either produce low quality results, or rely on additional priors such as target topology, appearance information, or input normal directions. In this paper, we present OReX, a method for 3D shape reconstruction from slices alone, featuring a Neural Field as the interpolation prior. A simple neural network is trained on the input planes to receive a 3D coordinate and return an inside/outside estimate for the query point. This prior is powerful in inducing smoothness and self-similarities. The main challenge for this approach is high-frequency details, as the neural prior is overly smoothing. To alleviate this, we offer an iterative estimation architecture and a hierarchical input sampling scheme that encourage coarse-to-fine training, allowing focusing on high frequencies at later stages. In addition, we identify and analyze a common ripple-like effect stemming from the mesh extraction step. We mitigate it by regularizing the spatial gradients of the indicator function around input in/out boundaries, cutting the problem at the root.

• PANeRF: Pseudo-view Augmentation for Improved Neural Radiance Fields Based on Few-shot Inputs | [code]

  The method of neural radiance fields (NeRF) has been developed in recent years, and this technology has promising applications for synthesizing novel views of complex scenes. However, NeRF requires dense input views, typically numbering in the hundreds, for generating high-quality images. With a decrease in the number of input views, the rendering quality of NeRF for unseen viewpoints tends to degenerate drastically. To overcome this challenge, we propose pseudo-view augmentation of NeRF, a scheme that expands a sufficient amount of data by considering the geometry of few-shot inputs. We first initialized the NeRF network by leveraging the expanded pseudo-views, which efficiently minimizes uncertainty when rendering unseen views. Subsequently, we fine-tuned the network by utilizing sparse-view inputs containing precise geometry and color information. Through experiments under various settings, we verified that our model faithfully synthesizes novel-view images of superior quality and outperforms existing methods for multi-view datasets.

• ActiveRMAP: Radiance Field for Active Mapping And Planning | [code]

  A high-quality 3D reconstruction of a scene from a collection of 2D images can be achieved through offline/online mapping methods. In this paper, we explore active mapping from the perspective of implicit representations, which have recently produced compelling results in a variety of applications. One of the most popular implicit representations - Neural Radiance Field (NeRF), first demonstrated photorealistic rendering results using multi-layer perceptrons, with promising offline 3D reconstruction as a by-product of the radiance field. More recently, researchers also applied this implicit representation for online reconstruction and localization (i.e. implicit SLAM systems). However, the study on using implicit representation for active vision tasks is still very limited. In this paper, we are particularly interested in applying the neural radiance field for active mapping and planning problems, which are closely coupled tasks in an active system. We, for the first time, present an RGB-only active vision framework using radiance field representation for active 3D reconstruction and planning in an online manner. Specifically, we formulate this joint task as an iterative dual-stage optimization problem, where we alternatively optimize for the radiance field representation and path planning. Experimental results suggest that the proposed method achieves competitive results compared to other offline methods and outperforms active reconstruction methods using NeRFs.

• Zero NeRF: Registration with Zero Overlap | [code]

  We present Zero-NeRF, a projective surface registration method that, to the best of our knowledge, offers the first general solution capable of alignment between scene representations with minimal or zero visual correspondence. To do this, we enforce consistency between visible surfaces of partial and complete reconstructions, which allows us to constrain occluded geometry. We use a NeRF as our surface representation and the NeRF rendering pipeline to perform this alignment. To demonstrate the efficacy of our method, we register real-world scenes from opposite sides with infinitesimal overlaps that cannot be accurately registered using prior methods, and we compare these results against widely used registration methods.

• FLNeRF: 3D Facial Landmarks Estimation in Neural Radiance Fields | [code]

  This paper presents the first significant work on directly predicting 3D face landmarks on neural radiance fields (NeRFs), without using intermediate representations such as 2D images, depth maps, or point clouds. Our 3D coarse-to-fine Face Landmarks NeRF (FLNeRF) model efficiently samples from the NeRF on the whole face with individual facial features for accurate landmarks. To mitigate the limited number of facial expressions in the available data, local and non-linear NeRF warp is applied at facial features in fine scale to simulate large emotions range, including exaggerated facial expressions (e.g., cheek blowing, wide opening mouth, eye blinking), for training FLNeRF. With such expression augmentation, our model can predict 3D landmarks not limited to the 20 discrete expressions given in the data. Robust 3D NeRF facial landmarks contribute to many downstream tasks. As an example, we modify MoFaNeRF to enable high-quality face editing and swapping using face landmarks on NeRF, allowing more direct control and wider range of complex expressions. Experiments show that the improved model using landmarks achieves comparable to better results.

• SPARF: Neural Radiance Fields from Sparse and Noisy Poses | [code]

  Neural Radiance Field (NeRF) has recently emerged as a powerful representation to synthesize photorealistic novel views. While showing impressive performance, it relies on the availability of dense input views with highly accurate camera poses, thus limiting its application in real-world scenarios. In this work, we introduce Sparse Pose Adjusting Radiance Field (SPARF), to address the challenge of novel-view synthesis given only few wide-baseline input images (as low as 3) with noisy camera poses. Our approach exploits multi-view geometry constraints in order to jointly learn the NeRF and refine the camera poses. By relying on pixel matches extracted between the input views, our multi-view correspondence objective enforces the optimized scene and camera poses to converge to a global and geometrically accurate solution. Our depth consistency loss further encourages the reconstructed scene to be consistent from any viewpoint. Our approach sets a new state of the art in the sparse-view regime on multiple challenging datasets.

• Tensor4D : Efficient Neural 4D Decomposition for High-fidelity Dynamic Reconstruction and Rendering | [code]

  We present Tensor4D, an efficient yet effective approach to dynamic scene modeling. The key of our solution is an efficient 4D tensor decomposition method so that the dynamic scene can be directly represented as a 4D spatio-temporal tensor. To tackle the accompanying memory issue, we decompose the 4D tensor hierarchically by projecting it first into three time-aware volumes and then nine compact feature planes. In this way, spatial information over time can be simultaneously captured in a compact and memory-efficient manner. When applying Tensor4D for dynamic scene reconstruction and rendering, we further factorize the 4D fields to different scales in the sense that structural motions and dynamic detailed changes can be learned from coarse to fine. The effectiveness of our method is validated on both synthetic and real-world scenes. Extensive experiments show that our method is able to achieve high-quality dynamic reconstruction and rendering from sparse-view camera rigs or even a monocular camera. The code and dataset will be released at this https URL.

• NeRF-RPN: A general framework for object detection in NeRFs | [code]

  This paper presents the first significant object detection framework, NeRF-RPN, which directly operates on NeRF. Given a pre-trained NeRF model, NeRF-RPN aims to detect all bounding boxes of objects in a scene. By exploiting a novel voxel representation that incorporates multi-scale 3D neural volumetric features, we demonstrate it is possible to regress the 3D bounding boxes of objects in NeRF directly without rendering the NeRF at any viewpoint. NeRF-RPN is a general framework and can be applied to detect objects without class labels. We experimented the NeRF-RPN with various backbone architectures, RPN head designs and loss functions. All of them can be trained in an end-to-end manner to estimate high quality 3D bounding boxes. To facilitate future research in object detection for NeRF, we built a new benchmark dataset which consists of both synthetic and real-world data with careful labeling and clean up. Please click this https URL for visualizing the 3D region proposals by our NeRF-RPN. Code and dataset will be made available.

• Local-to-Global Registration for Bundle-Adjusting Neural Radiance Fields | [code]

  Neural Radiance Fields (NeRF) have achieved photorealistic novel views synthesis; however, the requirement of accurate camera poses limits its application. Despite analysis-by-synthesis extensions for jointly learning neural 3D representations and registering camera frames exist, they are susceptible to suboptimal solutions if poorly initialized. We propose L2G-NeRF, a Local-to-Global registration method for bundle-adjusting Neural Radiance Fields: first, a pixel-wise flexible alignment, followed by a frame-wise constrained parametric alignment. Pixel-wise local alignment is learned in an unsupervised way via a deep network which optimizes photometric reconstruction errors. Frame-wise global alignment is performed using differentiable parameter estimation solvers on the pixel-wise correspondences to find a global transformation. Experiments on synthetic and real-world data show that our method outperforms the current state-of-the-art in terms of high-fidelity reconstruction and resolving large camera pose misalignment. Our module is an easy-to-use plugin that can be applied to NeRF variants and other neural field applications. The Code and supplementary materials are available at this https URL.

• SegNeRF: 3D Part Segmentation with Neural Radiance Fields | [code]

  Recent advances in Neural Radiance Fields (NeRF) boast impressive performances for generative tasks such as novel view synthesis and 3D reconstruction. Methods based on neural radiance fields are able to represent the 3D world implicitly by relying exclusively on posed images. Yet, they have seldom been explored in the realm of discriminative tasks such as 3D part segmentation. In this work, we attempt to bridge that gap by proposing SegNeRF: a neural field representation that integrates a semantic field along with the usual radiance field. SegNeRF inherits from previous works the ability to perform novel view synthesis and 3D reconstruction, and enables 3D part segmentation from a few images. Our extensive experiments on PartNet show that SegNeRF is capable of simultaneously predicting geometry, appearance, and semantic information from posed images, even for unseen objects. The predicted semantic fields allow SegNeRF to achieve an average mIoU of 30.30% for 2D novel view segmentation, and 37.46% for 3D part segmentation, boasting competitive performance against point-based methods by using only a few posed images. Additionally, SegNeRF is able to generate an explicit 3D model from a single image of an object taken in the wild, with its corresponding part segmentation.

• Shape, Pose, and Appearance from a Single Image via Bootstrapped Radiance Field Inversion | [code]

  Neural Radiance Fields (NeRF) coupled with GANs represent a promising direction in the area of 3D reconstruction from a single view, owing to their ability to efficiently model arbitrary topologies. Recent work in this area, however, has mostly focused on synthetic datasets where exact ground-truth poses are known, and has overlooked pose estimation, which is important for certain downstream applications such as augmented reality (AR) and robotics. We introduce a principled end-to-end reconstruction framework for natural images, where accurate ground-truth poses are not available. Our approach recovers an SDF-parameterized 3D shape, pose, and appearance from a single image of an object, without exploiting multiple views during training. More specifically, we leverage an unconditional 3D-aware generator, to which we apply a hybrid inversion scheme where a model produces a first guess of the solution which is then refined via optimization. Our framework can de-render an image in as few as 10 steps, enabling its use in practical scenarios. We demonstrate state-of-the-art results on a variety of real and synthetic benchmarks.

• Recovering Fine Details for Neural Implicit Surface Reconstruction | [code]

  Recent works on implicit neural representations have made significant strides. Learning implicit neural surfaces using volume rendering has gained popularity in multi-view reconstruction without 3D supervision. However, accurately recovering fine details is still challenging, due to the underlying ambiguity of geometry and appearance representation. In this paper, we present D-NeuS, a volume rendering-base neural implicit surface reconstruction method capable to recover fine geometry details, which extends NeuS by two additional loss functions targeting enhanced reconstruction quality. First, we encourage the rendered surface points from alpha compositing to have zero signed distance values, alleviating the geometry bias arising from transforming SDF to density for volume rendering. Second, we impose multi-view feature consistency on the surface points, derived by interpolating SDF zero-crossings from sampled points along rays. Extensive quantitative and qualitative results demonstrate that our method reconstructs high-accuracy surfaces with details, and outperforms the state of the art.

• Neural Puppeteer: Keypoint-Based Neural Rendering of Dynamic Shapes, ACCV2022 | [code]

  We introduce Neural Puppeteer, an efficient neural rendering pipeline for articulated shapes. By inverse rendering, we can predict 3D keypoints from multi-view 2D silhouettes alone, without requiring texture information. Furthermore, we can easily predict 3D keypoints of the same class of shapes with one and the same trained model and generalize more easily from training with synthetic data which we demonstrate by successfully applying zero-shot synthetic to real-world experiments. We demonstrate the flexibility of our method by fitting models to synthetic videos of different animals and a human, and achieve quantitative results which outperform our baselines. Our method uses 3D keypoints in conjunction with individual local feature vectors and a global latent code to allow for an efficient representation of time-varying and articulated shapes such as humans and animals. In contrast to previous work, we do not perform reconstruction in the 3D domain, but project the 3D features into 2D cameras and perform reconstruction of 2D RGB-D images from these projected features, which is significantly faster than volumetric rendering. Our synthetic dataset will be publicly available, to further develop the evolving field of animal pose and shape reconstruction.

• DynIBaR: Neural Dynamic Image-Based Rendering, - | [code]

  We address the problem of synthesizing novel views from a monocular video depicting a complex dynamic scene. State-of-the-art methods based on temporally varying Neural Radiance Fields (aka dynamic NeRFs) have shown impressive results on this task. However, for long videos with complex object motions and uncontrolled camera trajectories, these methods can produce blurry or inaccurate renderings, hampering their use in real-world applications. Instead of encoding the entire dynamic scene within the weights of an MLP, we present a new approach that addresses these limitations by adopting a volumetric image-based rendering framework that synthesizes new viewpoints by aggregating features from nearby views in a scene-motion-aware manner. Our system retains the advantages of prior methods in its ability to model complex scenes and view-dependent effects, but also enables synthesizing photo-realistic novel views from long videos featuring complex scene dynamics with unconstrained camera trajectories. We demonstrate significant improvements over state-of-the-art methods on dynamic scene datasets, and also apply our approach to in-the-wild videos with challenging camera and object motion, where prior methods fail to produce high-quality renderings. Our project webpage is at this http URL.

• Sampling Neural Radiance Fields for Refractive Objects, SIGGRAPH-Asia2022 | [code]

  Recently, differentiable volume rendering in neural radiance fields (NeRF) has gained a lot of popularity, and its variants have attained many impressive results. However, existing methods usually assume the scene is a homogeneous volume so that a ray is cast along the straight path. In this work, the scene is instead a heterogeneous volume with a piecewise-constant refractive index, where the path will be curved if it intersects the different refractive indices. For novel view synthesis of refractive objects, our NeRF-based framework aims to optimize the radiance fields of bounded volume and boundary from multi-view posed images with refractive object silhouettes. To tackle this challenging problem, the refractive index of a scene is reconstructed from silhouettes. Given the refractive index, we extend the stratified and hierarchical sampling techniques in NeRF to allow drawing samples along a curved path tracked by the Eikonal equation. The results indicate that our framework outperforms the state-of-the-art method both quantitatively and qualitatively, demonstrating better performance on the perceptual similarity metric and an apparent improvement in the rendering quality on several synthetic and real scenes.

Nov13 - Nov19, 2022

• Real-Time Omnidirectional Roaming in Large Scale Indoor Scenes, SIGGRAPH-Asia2022 | [code]

  Neural radiance field (NeRF) has recently achieved impressive results in novel view synthesis. However, previous works on NeRF mainly focus on object-centric scenarios. They would suffer observable performance degradation in outward-facing and large-scale scenes due to limiting positional encoding capacity. To narrow the gap, we explore radiance fields in a geometry-aware fashion. We estimate explicit geometry from the omnidirectional neural radiance field that was learned from multiple 360° images. Relying on the recovered geometry, we use an adaptive divide-and-conquer strategy to slim and fine-tune the radiance fields and further improve render speed and quality. Quantitative and qualitative comparisons among baselines illustrated our predominant performance in large-scale indoor scenes and our system supports real-time VR roaming.

• Magic3D: High-Resolution Text-to-3D Content Creation | [code]

  DreamFusion has recently demonstrated the utility of a pre-trained text-to-image diffusion model to optimize Neural Radiance Fields (NeRF), achieving remarkable text-to-3D synthesis results. However, the method has two inherent limitations: (a) extremely slow optimization of NeRF and (b) low-resolution image space supervision on NeRF, leading to low-quality 3D models with a long processing time. In this paper, we address these limitations by utilizing a two-stage optimization framework. First, we obtain a coarse model using a low-resolution diffusion prior and accelerate with a sparse 3D hash grid structure. Using the coarse representation as the initialization, we further optimize a textured 3D mesh model with an efficient differentiable renderer interacting with a high-resolution latent diffusion model. Our method, dubbed Magic3D, can create high quality 3D mesh models in 40 minutes, which is 2x faster than DreamFusion (reportedly taking 1.5 hours on average), while also achieving higher resolution. User studies show 61.7% raters to prefer our approach over DreamFusion. Together with the image-conditioned generation capabilities, we provide users with new ways to control 3D synthesis, opening up new avenues to various creative applications.

• AligNeRF: High-Fidelity Neural Radiance Fields via Alignment-Aware Training | [code]

  Neural Radiance Fields (NeRFs) are a powerful representation for modeling a 3D scene as a continuous function. Though NeRF is able to render complex 3D scenes with view-dependent effects, few efforts have been devoted to exploring its limits in a high-resolution setting. Specifically, existing NeRF-based methods face several limitations when reconstructing high-resolution real scenes, including a very large number of parameters, misaligned input data, and overly smooth details. In this work, we conduct the first pilot study on training NeRF with high-resolution data and propose the corresponding solutions: 1) marrying the multilayer perceptron (MLP) with convolutional layers which can encode more neighborhood information while reducing the total number of parameters; 2) a novel training strategy to address misalignment caused by moving objects or small camera calibration errors; and 3) a high-frequency aware loss. Our approach is nearly free without introducing obvious training/testing costs, while experiments on different datasets demonstrate that it can recover more high-frequency details compared with the current state-of-the-art NeRF models. Project page: \url{this https URL.}

• 3DLatNav: Navigating Generative Latent Spaces for Semantic-Aware 3D Object Manipulation | [code]

  3D generative models have been recently successful in generating realistic 3D objects in the form of point clouds. However, most models do not offer controllability to manipulate the shape semantics of component object parts without extensive semantic attribute labels or other reference point clouds. Moreover, beyond the ability to perform simple latent vector arithmetic or interpolations, there is a lack of understanding of how part-level semantics of 3D shapes are encoded in their corresponding generative latent spaces. In this paper, we propose 3DLatNav; a novel approach to navigating pretrained generative latent spaces to enable controlled part-level semantic manipulation of 3D objects. First, we propose a part-level weakly-supervised shape semantics identification mechanism using latent representations of 3D shapes. Then, we transfer that knowledge to a pretrained 3D object generative latent space to unravel disentangled embeddings to represent different shape semantics of component parts of an object in the form of linear subspaces, despite the unavailability of part-level labels during the training. Finally, we utilize those identified subspaces to show that controllable 3D object part manipulation can be achieved by applying the proposed framework to any pretrained 3D generative model. With two novel quantitative metrics to evaluate the consistency and localization accuracy of part-level manipulations, we show that 3DLatNav outperforms existing unsupervised latent disentanglement methods in identifying latent directions that encode part-level shape semantics of 3D objects. With multiple ablation studies and testing on state-of-the-art generative models, we show that 3DLatNav can implement controlled part-level semantic manipulations on an input point cloud while preserving other features and the realistic nature of the object.

• RenderDiffusion: Image Diffusion for 3D Reconstruction, Inpainting and Generation | [code]

  Diffusion models currently achieve state-of-the-art performance for both conditional and unconditional image generation. However, so far, image diffusion models do not support tasks required for 3D understanding, such as view-consistent 3D generation or single-view object reconstruction. In this paper, we present RenderDiffusion as the first diffusion model for 3D generation and inference that can be trained using only monocular 2D supervision. At the heart of our method is a novel image denoising architecture that generates and renders an intermediate three-dimensional representation of a scene in each denoising step. This enforces a strong inductive structure into the diffusion process that gives us a 3D consistent representation while only requiring 2D supervision. The resulting 3D representation can be rendered from any viewpoint. We evaluate RenderDiffusion on ShapeNet and Clevr datasets and show competitive performance for generation of 3D scenes and inference of 3D scenes from 2D images. Additionally, our diffusion-based approach allows us to use 2D inpainting to edit 3D scenes. We believe that our work promises to enable full 3D generation at scale when trained on massive image collections, thus circumventing the need to have large-scale 3D model collections for supervision.

• DINER: Disorder-Invariant Implicit Neural Representation | [code]

  Implicit neural representation (INR) characterizes the attributes of a signal as a function of corresponding coordinates which emerges as a sharp weapon for solving inverse problems. However, the capacity of INR is limited by the spectral bias in the network training. In this paper, we find that such a frequency-related problem could be largely solved by re-arranging the coordinates of the input signal, for which we propose the disorder-invariant implicit neural representation (DINER) by augmenting a hash-table to a traditional INR backbone. Given discrete signals sharing the same histogram of attributes and different arrangement orders, the hash-table could project the coordinates into the same distribution for which the mapped signal can be better modeled using the subsequent INR network, leading to significantly alleviated spectral bias. Experiments not only reveal the generalization of the DINER for different INR backbones (MLP vs. SIREN) and various tasks (image/video representation, phase retrieval, and refractive index recovery) but also show the superiority over the state-of-the-art algorithms both in quality and speed.

• Latent-NeRF for Shape-Guided Generation of 3D Shapes and Textures | [code]

  Text-guided image generation has progressed rapidly in recent years, inspiring major breakthroughs in text-guided shape generation. Recently, it has been shown that using score distillation, one can successfully text-guide a NeRF model to generate a 3D object. We adapt the score distillation to the publicly available, and computationally efficient, Latent Diffusion Models, which apply the entire diffusion process in a compact latent space of a pretrained autoencoder. As NeRFs operate in image space, a naive solution for guiding them with latent score distillation would require encoding to the latent space at each guidance step. Instead, we propose to bring the NeRF to the latent space, resulting in a Latent-NeRF. Analyzing our Latent-NeRF, we show that while Text-to-3D models can generate impressive results, they are inherently unconstrained and may lack the ability to guide or enforce a specific 3D structure. To assist and direct the 3D generation, we propose to guide our Latent-NeRF using a Sketch-Shape: an abstract geometry that defines the coarse structure of the desired object. Then, we present means to integrate such a constraint directly into a Latent-NeRF. This unique combination of text and shape guidance allows for increased control over the generation process. We also show that latent score distillation can be successfully applied directly on 3D meshes. This allows for generating high-quality textures on a given geometry. Our experiments validate the power of our different forms of guidance and the efficiency of using latent rendering. Implementation is available at this https URL

• AsyncNeRF: Learning Large-scale Radiance Fields from Asynchronous RGB-D Sequences with Time-Pose Function | [code]

  Large-scale radiance fields are promising mapping tools for smart transportation applications like autonomous driving or drone delivery. But for large-scale scenes, compact synchronized RGB-D cameras are not applicable due to limited sensing range, and using separate RGB and depth sensors inevitably leads to unsynchronized sequences. Inspired by the recent success of self-calibrating radiance field training methods that do not require known intrinsic or extrinsic parameters, we propose the first solution that self-calibrates the mismatch between RGB and depth frames. We leverage the important domain-specific fact that RGB and depth frames are actually sampled from the same trajectory and develop a novel implicit network called the time-pose function. Combining it with a large-scale radiance field leads to an architecture that cascades two implicit representation networks. To validate its effectiveness, we construct a diverse and photorealistic dataset that covers various RGB-D mismatch scenarios. Through a comprehensive benchmarking on this dataset, we demonstrate the flexibility of our method in different scenarios and superior performance over applicable prior counterparts. Codes, data, and models will be made publicly available.

Nov6 - Nov12, 2022

• NeXT: Towards High Quality Neural Radiance Fields via Multi-skip Transformer, ECCV2022 | [code]

  Neural Radiance Fields (NeRF) methods show impressive performance for novel view synthesis by representing a scene via a neural network. However, most existing NeRF based methods, including its variants, treat each sample point individually as input, while ignoring the inherent relationships between adjacent sample points from the corresponding rays, thus hindering the reconstruction performance. To address this issue, we explore a brand new scheme, namely NeXT, introducing a multi-skip transformer to capture the rich relationships between various sample points in a ray-level query. Specifically, ray tokenization is proposed to represent each ray as a sequence of point embeddings which is taken as input of our proposed NeXT. In this way, relationships between sample points are captured via the built-in self-attention mechanism to promote the reconstruction. Besides, our proposed NeXT can be easily combined with other NeRF based methods to improve their rendering quality. Extensive experiments conducted on three datasets demonstrate that NeXT significantly outperforms all previous state-of-the-art work by a large margin. In particular, the proposed NeXT surpasses the strong NeRF baseline by 2.74 dB of PSNR on Blender dataset. The code is available at https://github.com/Crishawy/NeXT.

• Directed Ray Distance Functions for 3D Scene Reconstruction, ECCV2022 | [code]

  We present an approach for full 3D scene reconstruction from a single unseen image. We trained on dataset of realistic non-watertight scans of scenes. Our approach uses a predicted distance function, since these have shown promise in handling complex topologies and large spaces. We identify and analyze two key challenges for predicting such image conditioned distance functions that have prevented their success on real 3D scene data. First, we show that predicting a conventional scene distance from an image requires reasoning over a large receptive field. Second, we analytically show that the optimal output of the network trained to predict these distance functions does not obey all the distance function properties. We propose an alternate distance function, the Directed Ray Distance Function (DRDF), that tackles both challenges. We show that a deep network trained to predict DRDFs outperforms all other methods quantitatively and qualitatively on 3D reconstruction from single image on Matterport3D, 3DFront, and ScanNet. (Project Page: https://nileshkulkarni.github.io/scene_drdf)

• ParticleNeRF: A Particle-Based Encoding for Online Neural Radiance Fields in Dynamic Scenes | [code]

  Neural Radiance Fields (NeRFs) learn implicit representations of - typically static - environments from images. Our paper extends NeRFs to handle dynamic scenes in an online fashion. We propose ParticleNeRF that adapts to changes in the geometry of the environment as they occur, learning a new up-to-date representation every 350 ms. ParticleNeRF can represent the current state of dynamic environments with much higher fidelity as other NeRF frameworks. To achieve this, we introduce a new particle-based parametric encoding, which allows the intermediate NeRF features - now coupled to particles in space - to move with the dynamic geometry. This is possible by backpropagating the photometric reconstruction loss into the position of the particles. The position gradients are interpreted as particle velocities and integrated into positions using a position-based dynamics (PBS) physics system. Introducing PBS into the NeRF formulation allows us to add collision constraints to the particle motion and creates future opportunities to add other movement priors into the system, such as rigid and deformable body

• Temporal Coherence-Based Distributed Ray Tracing of Massive Scenes, ToG2022 | [code]

  Distributed ray tracing algorithms are widely used when rendering massive scenes, where data utilization and load balancing are the keys to improving performance. One essential observation is that rays are temporally coherent, which indicates that temporal information can be used to improve computational efficiency. In this paper, we use temporal coherence to optimize the performance of distributed ray tracing. First, we propose a temporal coherence-based scheduling algorithm to guide the task/data assignment and scheduling. Then, we propose a virtual portal structure to predict the radiance of rays based on the previous frame, and send the rays with low radiance to a precomputed simplified model for further tracing, which can dramatically reduce the traversal complexity and the overhead of network data transmission. The approach was validated on scenes of sizes up to 355 GB. Our algorithm can achieve a speedup of up to 81% compared to previous algorithms, with a very small mean squared error.

• QRF: Implicit Neural Representations with Quantum Radiance Fields | [code]

  Photorealistic rendering of real-world scenes is a tremendous challenge with a wide range of applications, including mixed reality (MR), and virtual reality (VR). Neural networks, which have long been investigated in the context of solving differential equations, have previously been introduced as implicit representations for photorealistic rendering. However, realistic rendering using classic computing is challenging because it requires time-consuming optical ray marching, and suffer computational bottlenecks due to the curse of dimensionality. In this paper, we propose Quantum Radiance Fields (QRF), which integrate the quantum circuit, quantum activation function, and quantum volume rendering for implicit scene representation. The results indicate that QRF not only exploits the advantage of quantum computing, such as high speed, fast convergence, and high parallelism, but also ensure high quality of volume rendering.

• Common Pets in 3D: Dynamic New-View Synthesis of Real-Life Deformable Categories | [code]

  Obtaining photorealistic reconstructions of objects from sparse views is inherently ambiguous and can only be achieved by learning suitable reconstruction priors. Earlier works on sparse rigid object reconstruction successfully learned such priors from large datasets such as CO3D. In this paper, we extend this approach to dynamic objects. We use cats and dogs as a representative example and introduce Common Pets in 3D (CoP3D), a collection of crowd-sourced videos showing around 4,200 distinct pets. CoP3D is one of the first large-scale datasets for benchmarking non-rigid 3D reconstruction "in the wild". We also propose Tracker-NeRF, a method for learning 4D reconstruction from our dataset. At test time, given a small number of video frames of an unseen object, Tracker-NeRF predicts the trajectories of its 3D points and generates new views, interpolating viewpoint and time. Results on CoP3D reveal significantly better non-rigid new-view synthesis performance than existing baselines.

• Learning-based Inverse Rendering of Complex Indoor Scenes with Differentiable Monte Carlo Raytracing, SIGGRAPH-Asia2022 | [code]

  We present a learning-based approach for inverse rendering of complex indoor scenes with differentiable Monte Carlo raytracing. Our method takes a single indoor scene RGB image as input and automatically infers its underlying surface reflectance , geometry, and spatially-varying illumination. This enables us to perform photorealistic editing of the scene, such as inserting multiple complex virtual objects and editing surface materials faithfully with global illumination.

Oct30 - Nov5, 2022

• Deep Appearance Prefiltering, ToG2022 | [code]

  Physically based rendering of complex scenes can be prohibitively costly with a potentially unbounded and uneven distribution of complexity across the rendered image. The goal of an ideal level of detail (LoD) method is to make rendering costs independent of the 3D scene complexity, while preserving the appearance of the scene. However, current prefiltering LoD methods are limited in the appearances they can support due to their reliance of approximate models and other heuristics. We propose the first comprehensive multi-scale LoD framework for prefiltering 3D environments with complex geometry and materials (e.g., the Disney BRDF), while maintaining the appearance with respect to the ray-traced reference. Using a multi-scale hierarchy of the scene, we perform a data-driven prefiltering step to obtain an appearance phase function and directional coverage mask at each scale. At the heart of our approach is a novel neural representation that encodes this information into a compact latent form that is easy to decode inside a physically based renderer. Once a scene is baked out, our method requires no original geometry, materials, or textures at render time. We demonstrate that our approach compares favorably to state-of-the-art prefiltering methods and achieves considerable savings in memory for complex scenes.

• Neural Grasp Distance Fields for Robot Manipulation | [code]

  We formulate grasp learning as a neural field and present Neural Grasp Distance Fields (NGDF). Here, the input is a 6D pose of a robot end effector and output is a distance to a continuous manifold of valid grasps for an object. In contrast to current approaches that predict a set of discrete candidate grasps, the distance-based NGDF representation is easily interpreted as a cost, and minimizing this cost produces a successful grasp pose. This grasp distance cost can be incorporated directly into a trajectory optimizer for joint optimization with other costs such as trajectory smoothness and collision avoidance. During optimization, as the various costs are balanced and minimized, the grasp target is allowed to smoothly vary, as the learned grasp field is continuous. In simulation benchmarks with a Franka arm, we find that joint grasping and planning with NGDF outperforms baselines by 63% execution success while generalizing to unseen query poses and unseen object shapes. Project page: this https URL.

• nerf2nerf: Pairwise Registration of Neural Radiance Fields | [code]

  We introduce a technique for pairwise registration of neural fields that extends classical optimization-based local registration (i.e. ICP) to operate on Neural Radiance Fields (NeRF) -- neural 3D scene representations trained from collections of calibrated images. NeRF does not decompose illumination and color, so to make registration invariant to illumination, we introduce the concept of a ''surface field'' -- a field distilled from a pre-trained NeRF model that measures the likelihood of a point being on the surface of an object. We then cast nerf2nerf registration as a robust optimization that iteratively seeks a rigid transformation that aligns the surface fields of the two scenes. We evaluate the effectiveness of our technique by introducing a dataset of pre-trained NeRF scenes -- our synthetic scenes enable quantitative evaluations and comparisons to classical registration techniques, while our real scenes demonstrate the validity of our technique in real-world scenarios. Additional results available at: this https URL

• HyperSound: Generating Implicit Neural Representations of Audio Signals with Hypernetworks | [code]

  Implicit neural representations (INRs) are a rapidly growing research field, which provides alternative ways to represent multimedia signals. Recent applications of INRs include image super-resolution, compression of high-dimensional signals, or 3D rendering. However, these solutions usually focus on visual data, and adapting them to the audio domain is not trivial. Moreover, it requires a separately trained model for every data sample. To address this limitation, we propose HyperSound, a meta-learning method leveraging hypernetworks to produce INRs for audio signals unseen at training time. We show that our approach can reconstruct sound waves with quality comparable to other state-of-the-art models.

• Attention-based Neural Cellular Automata, NeurIPS2022 | [code]

  Recent extensions of Cellular Automata (CA) have incorporated key ideas from modern deep learning, dramatically extending their capabilities and catalyzing a new family of Neural Cellular Automata (NCA) techniques. Inspired by Transformer-based architectures, our work presents a new class of attention-based NCAs formed using a spatially localized—yet globally organized—self-attention scheme. We introduce an instance of this class named Vision Transformer Cellular Automata (ViTCA). We present quantitative and qualitative results on denoising autoencoding across six benchmark datasets, comparing ViTCA to a U-Net, a U-Net-based CA baseline (UNetCA), and a Vision Transformer (ViT). When comparing across architectures configured to similar parameter complexity, ViTCA architectures yield superior performance across all benchmarks and for nearly every evaluation metric. We present an ablation study on various architectural configurations of ViTCA, an analysis of its effect on cell states, and an investigation on its inductive biases. Finally, we examine its learned representations via linear probes on its converged cell state hidden representations, yielding, on average, superior results when compared to our U-Net, ViT, and UNetCA baselines.

• GARF: Gaussian Activated Radiance Fields for High Fidelity Reconstruction and Pose Estimation, ECCV2022 | [code]

  Despite Neural Radiance Fields (NeRF) showing compelling results in photorealistic novel views synthesis of real-world scenes, most existing approaches require accurate prior camera poses. Although approaches for jointly recovering the radiance field and camera pose exist (BARF), they rely on a cumbersome coarse-to-fine auxiliary positional embedding to ensure good performance. We present Gaussian Activated neural Radiance Fields (GARF), a new positional embedding-free neural radiance field architecture - employing Gaussian activations - that outperforms the current state-of-the-art in terms of high fidelity reconstruction and pose estimation.

• Learning Neural Implicit Representations with Surface Signal Parameterizations | [code]

  Neural implicit surface representations have recently emerged as popular alternative to explicit 3D object encodings, such as polygonal meshes, tabulated points, or voxels. While significant work has improved the geometric fidelity of these representations, much less attention is given to their final appearance. Traditional explicit object representations commonly couple the 3D shape data with auxiliary surface-mapped image data, such as diffuse color textures and fine-scale geometric details in normal maps that typically require a mapping of the 3D surface onto a plane, i.e., a surface parameterization; implicit representations, on the other hand, cannot be easily textured due to lack of configurable surface parameterization. Inspired by this digital content authoring methodology, we design a neural network architecture that implicitly encodes the underlying surface parameterization suitable for appearance data. As such, our model remains compatible with existing mesh-based digital content with appearance data. Motivated by recent work that overfits compact networks to individual 3D objects, we present a new weight-encoded neural implicit representation that extends the capability of neural implicit surfaces to enable various common and important applications of texture mapping. Our method outperforms reasonable baselines and state-of-the-art alternatives.

• gCoRF: Generative Compositional Radiance Fields, 3DV2022 | [code]

  3D generative models of objects enable photorealistic image synthesis with 3D control. Existing methods model the scene as a global scene representation, ignoring the compositional aspect of the scene. Compositional reasoning can enable a wide variety of editing applications, in addition to enabling generalizable 3D reasoning. In this paper, we present a compositional generative model, where each semantic part of the object is represented as an independent 3D representation learnt from only in-the-wild 2D data. We start with a global generative model (GAN) and learn to decompose it into different semantic parts using supervision from 2D segmentation masks. We then learn to composite independently sampled parts in order to create coherent global scenes. Different parts can be independently sampled, while keeping rest of the object fixed. We evaluate our method on a wide variety of objects and parts, and demonstrate editing applications.

• Digging into Radiance Grid for Real-Time View Synthesis with Detail Preservation, ECCV2022 | [code]

  Neural Radiance Fields (NeRF) [31] series are impressive in representing scenes and synthesizing high-quality novel views. However, most previous works fail to preserve texture details and suffer from slow training speed. A recent method SNeRG [11] demonstrates that baking a trained NeRF as a Sparse Neural Radiance Grid enables real-time view synthesis with slight scarification of rendering quality. In this paper, we dig into the Radiance Grid representation and present a set of improvements, which together result in boosted performance in terms of both speed and quality. First, we propose an HieRarchical Sparse Radiance Grid (HrSRG) representation that has higher voxel resolution for informative spaces and fewer voxels for other spaces. HrSRG leverages a hierarchical voxel grid building process inspired by [30, 55], and can describe a scene at high resolution without excessive memory footprint. Furthermore, we show that directly optimizing the voxel grid leads to surprisingly good texture details in rendered images. This direct optimization is memory-friendly and requires multiple orders of magnitude less time than conventional NeRFs as it only involves a tiny MLP. Finally, we find that a critical factor that prevents fine details restoration is the misaligned 2D pixels among images caused by camera pose errors. We propose to use the perceptual loss to add tolerance to misalignments, leading to the improved visual quality of rendered images.

Oct23 - Oct29, 2022

• NeX360: Real-time All-around View Synthesis with Neural Basis Expansion, TPAMI2022 | [code]

  We present NeX, a new approach to novel view synthesis based on enhancements of multiplane images (MPI) that can reproduce view-dependent effects in real time. Unlike traditional MPI, our technique parameterizes each pixel as a linear combination of spherical basis functions learned from a neural network to model view-dependent effects and uses a hybrid implicit-explicit modeling strategy to improve fine detail. Moreover, we also present an extension to NeX, which leverages knowledge distillation to train multiple MPIs for unbounded 360 ∘ scenes. Our method is evaluated on several benchmark datasets: NeRF-Synthetic dataset, Light Field dataset, Real Forward-Facing dataset, Space dataset, as well as Shiny , our new dataset that contains significantly more challenging view-dependent effects, such as the rainbow reflections on the CD. Our method outperforms other real-time rendering approaches on PSNR, SSIM, and LPIPS and can render unbounded 360 ∘ scenes in real time.

• NeRFPlayer: A Streamable Dynamic Scene Representation with Decomposed Neural Radiance Fields | [code]

  Visually exploring in a real-world 4D spatiotemporal space freely in VR has been a long-term quest. The task is especially appealing when only a few or even single RGB cameras are used for capturing the dynamic scene. To this end, we present an efficient framework capable of fast reconstruction, compact modeling, and streamable rendering. First, we propose to decompose the 4D spatiotemporal space according to temporal characteristics. Points in the 4D space are associated with probabilities of belonging to three categories: static, deforming, and new areas. Each area is represented and regularized by a separate neural field. Second, we propose a hybrid representations based feature streaming scheme for efficiently modeling the neural fields. Our approach, coined NeRFPlayer, is evaluated on dynamic scenes captured by single hand-held cameras and multi-camera arrays, achieving comparable or superior rendering performance in terms of quality and speed comparable to recent state-of-the-art methods, achieving reconstruction in 10 seconds per frame and real-time rendering.

• Vox-Fusion: Dense Tracking and Mapping with Voxel-based Neural Implicit Representation | [code]

  In this work, we present a dense tracking and mapping system named Vox-Fusion, which seamlessly fuses neural implicit representations with traditional volumetric fusion methods. Our approach is inspired by the recently developed implicit mapping and positioning system and further extends the idea so that it can be freely applied to practical scenarios. Specifically, we leverage a voxel-based neural implicit surface representation to encode and optimize the scene inside each voxel. Furthermore, we adopt an octree-based structure to divide the scene and support dynamic expansion, enabling our system to track and map arbitrary scenes without knowing the environment like in previous works. Moreover, we proposed a high-performance multi-process framework to speed up the method, thus supporting some applications that require real-time performance. The evaluation results show that our methods can achieve better accuracy and completeness than previous methods. We also show that our Vox-Fusion can be used in augmented reality and virtual reality applications. Our source code is publicly available at this https URL.

• Boosting Point Clouds Rendering via Radiance Mapping | [code]

  Recent years we have witnessed rapid development in NeRF-based image rendering due to its high quality. However, point clouds rendering is somehow less explored. Compared to NeRF-based rendering which suffers from dense spatial sampling, point clouds rendering is naturally less computation intensive, which enables its deployment in mobile computing device. In this work, we focus on boosting the image quality of point clouds rendering with a compact model design. We first analyze the adaption of the volume rendering formulation on point clouds. Based on the analysis, we simplify the NeRF representation to a spatial mapping function which only requires single evaluation per pixel. Further, motivated by ray marching, we rectify the the noisy raw point clouds to the estimated intersection between rays and surfaces as queried coordinates, which could avoid spatial frequency collapse and neighbor point disturbance. Composed of rasterization, spatial mapping and the refinement stages, our method achieves the state-of-the-art performance on point clouds rendering, outperforming prior works by notable margins, with a smaller model size. We obtain a PSNR of 31.74 on NeRF-Synthetic, 25.88 on ScanNet and 30.81 on DTU. Code and data would be released soon.

• Streaming Radiance Fields for 3D Video Synthesis, NeurIPS2022 | [code]

  We present an explicit-grid based method for efficiently reconstructing streaming radiance fields for novel view synthesis of real world dynamic scenes. Instead of training a single model that combines all the frames, we formulate the dynamic modeling problem with an incremental learning paradigm in which per-frame model difference is trained to complement the adaption of a base model on the current frame. By exploiting the simple yet effective tuning strategy with narrow bands, the proposed method realizes a feasible framework for handling video sequences on-the-fly with high training efficiency. The storage overhead induced by using explicit grid representations can be significantly reduced through the use of model difference based compression. We also introduce an efficient strategy to further accelerate model optimization for each frame. Experiments on challenging video sequences demonstrate that our approach is capable of achieving a training speed of 15 seconds per-frame with competitive rendering quality, which attains 1000× speedup over the state-of-the-art implicit methods. Code is available at this https URL.

• EpipolarNVS: leveraging on Epipolar geometry for single-image Novel View Synthesis, BMVC2022 | [code]

  Novel-view synthesis (NVS) can be tackled through different approaches, depending on the general setting: a single source image to a short video sequence, exact or noisy camera pose information, 3D-based information such as point clouds etc. The most challenging scenario, the one where we stand in this work, only considers a unique source image to generate a novel one from another viewpoint. However, in such a tricky situation, the latest learning-based solutions often struggle to integrate the camera viewpoint transformation. Indeed, the extrinsic information is often passed as-is, through a low-dimensional vector. It might even occur that such a camera pose, when parametrized as Euler angles, is quantized through a one-hot representation. This vanilla encoding choice prevents the learnt architecture from inferring novel views on a continuous basis (from a camera pose perspective). We claim it exists an elegant way to better encode relative camera pose, by leveraging 3D-related concepts such as the epipolar constraint. We, therefore, introduce an innovative method that encodes the viewpoint transformation as a 2D feature image. Such a camera encoding strategy gives meaningful insights to the network regarding how the camera has moved in space between the two views. By encoding the camera pose information as a finite number of coloured epipolar lines, we demonstrate through our experiments that our strategy outperforms vanilla encoding.

• NeRF-SLAM: Real-Time Dense Monocular SLAM with Neural Radiance Fields | [code]

  We propose a novel geometric and photometric 3D mapping pipeline for accurate and real-time scene reconstruction from monocular images. To achieve this, we leverage recent advances in dense monocular SLAM and real-time hierarchical volumetric neural radiance fields. Our insight is that dense monocular SLAM provides the right information to fit a neural radiance field of the scene in real-time, by providing accurate pose estimates and depth-maps with associated uncertainty. With our proposed uncertainty-based depth loss, we achieve not only good photometric accuracy, but also great geometric accuracy. In fact, our proposed pipeline achieves better geometric and photometric accuracy than competing approaches (up to 179% better PSNR and 86% better L1 depth), while working in real-time and using only monocular images.

• Compressing Explicit Voxel Grid Representations: fast NeRFs become also small | [code]

  NeRFs have revolutionized the world of per-scene radiance field reconstruction because of their intrinsic compactness. One of the main limitations of NeRFs is their slow rendering speed, both at training and inference time. Recent research focuses on the optimization of an explicit voxel grid (EVG) that represents the scene, which can be paired with neural networks to learn radiance fields. This approach significantly enhances the speed both at train and inference time, but at the cost of large memory occupation. In this work we propose Re:NeRF, an approach that specifically targets EVG-NeRFs compressibility, aiming to reduce memory storage of NeRF models while maintaining comparable performance. We benchmark our approach with three different EVG-NeRF architectures on four popular benchmarks, showing Re:NeRF's broad usability and effectiveness.

Oct16 - Oct22, 2022

• Compressing multidimensional weather and climate data into neural networks | [code]

  Weather and climate simulations produce petabytes of high-resolution data that are later analyzed by researchers in order to understand climate change or severe weather. We propose a new method of compressing this multidimensional weather and climate data: a coordinate-based neural network is trained to overfit the data, and the resulting parameters are taken as a compact representation of the original grid-based data. While compression ratios range from 300x to more than 3,000x, our method outperforms the state-of-the-art compressor SZ3 in terms of weighted RMSE, MAE. It can faithfully preserve important large scale atmosphere structures and does not introduce artifacts. When using the resulting neural network as a 790x compressed dataloader to train the WeatherBench forecasting model, its RMSE increases by less than 2%. The three orders of magnitude compression democratizes access to high-resolution climate data and enables numerous new research directions.

• NeARportation: A Remote Real-time Neural Rendering Framework, VRST22 | [code]

  While the presentation of photo-realistic appearance plays a major role in immersion in an augmented virtuality environment, displaying the photo-realistic appearance of real objects remains a challenging problem. Recent developments in photogrammetry have facilitated the incorporation of real objects into virtual space. However, photo-realistic photogrammetry requires a dedicated measurement environment, and there is a trade-off between measurement cost and quality. Furthermore, even with photo-realistic appearance measurements, there is a trade-off between rendering quality and framerate. There is no framework that could resolve these trade-offs and easily provide a photo-realistic appearance in real-time. Our NeARportation framework combines server-client bidirectional communication and neural rendering to resolve these trade-offs. Neural rendering on the server receives the client's head posture and generates a novel-view image with realistic appearance reproduction, which is streamed onto the client's display. By applying our framework to a stereoscopic display, we confirmed that it could display a high-fidelity appearance on full-HD stereo videos at 35-40 frames-per-second (fps), according to the user's head motion.

• Neural Sound Field Decomposition with Super-resolution of Sound Direction | [code]

  Sound field decomposition predicts waveforms in arbitrary directions using signals from a limited number of microphones as inputs. Sound field decomposition is fundamental to downstream tasks, including source localization, source separation, and spatial audio reproduction. Conventional sound field decomposition methods such as Ambisonics have limited spatial decomposition resolution. This paper proposes a learning-based Neural Sound field Decomposition (NeSD) framework to allow sound field decomposition with fine spatial direction resolution, using recordings from microphone capsules of a few microphones at arbitrary positions. The inputs of a NeSD system include microphone signals, microphone positions, and queried directions. The outputs of a NeSD include the waveform and the presence probability of a queried position. We model the NeSD systems respectively with different neural networks, including fully connected, time delay, and recurrent neural networks. We show that the NeSD systems outperform conventional Ambisonics and DOANet methods in sound field decomposition and source localization on speech, music, and sound events datasets. Demos are available at this https URL.

• An Exploration of Neural Radiance Field Scene Reconstruction: Synthetic, Real-world and Dynamic Scenes | [code]

  This project presents an exploration into 3D scene reconstruction of synthetic and real-world scenes using Neural Radiance Field (NeRF) approaches. We primarily take advantage of the reduction in training and rendering time of neural graphic primitives multi-resolution hash encoding, to reconstruct static video game scenes and real-world scenes, comparing and observing reconstruction detail and limitations. Additionally, we explore dynamic scene reconstruction using Neural Radiance Fields for Dynamic Scenes(D-NeRF). Finally, we extend the implementation of D-NeRF, originally constrained to handle synthetic scenes to also handle real-world dynamic scenes.

• Generative Range Imaging for Learning Scene Priors of 3D LiDAR Data, WACV2023 | [code]

  3D LiDAR sensors are indispensable for the robust vision of autonomous mobile robots. However, deploying LiDAR-based perception algorithms often fails due to a domain gap from the training environment, such as inconsistent angular resolution and missing properties. Existing studies have tackled the issue by learning inter-domain mapping, while the transferability is constrained by the training configuration and the training is susceptible to peculiar lossy noises called ray-drop. To address the issue, this paper proposes a generative model of LiDAR range images applicable to the data-level domain transfer. Motivated by the fact that LiDAR measurement is based on point-by-point range imaging, we train an implicit image representation-based generative adversarial networks along with a differentiable ray-drop effect. We demonstrate the fidelity and diversity of our model in comparison with the point-based and image-based state-of-the-art generative models. We also showcase upsampling and restoration applications. Furthermore, we introduce a Sim2Real application for LiDAR semantic segmentation. We demonstrate that our method is effective as a realistic ray-drop simulator and outperforms state-of-the-art methods.

• HDHumans: A Hybrid Approach for High-fidelity Digital Humans | [code]

  Photo-real digital human avatars are of enormous importance in graphics, as they enable immersive communication over the globe, improve gaming and entertainment experiences, and can be particularly beneficial for AR and VR settings. However, current avatar generation approaches either fall short in high-fidelity novel view synthesis, generalization to novel motions, reproduction of loose clothing, or they cannot render characters at the high resolution offered by modern displays. To this end, we propose HDHumans, which is the first method for HD human character synthesis that jointly produces an accurate and temporally coherent 3D deforming surface and highly photo-realistic images of arbitrary novel views and of motions not seen at training time. At the technical core, our method tightly integrates a classical deforming character template with neural radiance fields (NeRF). Our method is carefully designed to achieve a synergy between classical surface deformation and NeRF. First, the template guides the NeRF, which allows synthesizing novel views of a highly dynamic and articulated character and even enables the synthesis of novel motions. Second, we also leverage the dense pointclouds resulting from NeRF to further improve the deforming surface via 3D-to-3D supervision. We outperform the state of the art quantitatively and qualitatively in terms of synthesis quality and resolution, as well as the quality of 3D surface reconstruction.

• High-Quality RGB-D Reconstruction via Multi-View Uncalibrated Photometric Stereo and Gradient-SDF, WACV2023 | [code]

  Fine-detailed reconstructions are in high demand in many applications. However, most of the existing RGB-D reconstruction methods rely on pre-calculated accurate camera poses to recover the detailed surface geometry, where the representation of a surface needs to be adapted when optimizing different quantities. In this paper, we present a novel multi-view RGB-D based reconstruction method that tackles camera pose, lighting, albedo, and surface normal estimation via the utilization of a gradient signed distance field (gradient-SDF). The proposed method formulates the image rendering process using specific physically-based model(s) and optimizes the surface's quantities on the actual surface using its volumetric representation, as opposed to other works which estimate surface quantities only near the actual surface. To validate our method, we investigate two physically-based image formation models for natural light and point light source applications. The experimental results on synthetic and real-world datasets demonstrate that the proposed method can recover high-quality geometry of the surface more faithfully than the state-of-the-art and further improves the accuracy of estimated camera poses.

• Neural Fields for Robotic Object Manipulation from a Single Image, ICRA2023 | [code]

  We present a unified and compact representation for object rendering, 3D reconstruction, and grasp pose prediction that can be inferred from a single image within a few seconds. We achieve this by leveraging recent advances in the Neural Radiance Field (NeRF) literature that learn category-level priors and fine-tune on novel objects with minimal data and time. Our insight is that we can learn a compact shape representation and extract meaningful additional information from it, such as grasping poses. We believe this to be the first work to retrieve grasping poses directly from a NeRF-based representation using a single viewpoint (RGB-only), rather than going through a secondary network and/or representation. When compared to prior art, our method is two to three orders of magnitude smaller while achieving comparable performance at view reconstruction and grasping. Accompanying our method, we also propose a new dataset of rendered shoes for training a sim-2-real NeRF method with grasping poses for different widths of grippers.

• TANGO: Text-driven Photorealistic and Robust 3D Stylization via Lighting Decomposition, NeurIPS2022 | [code]

  Creation of 3D content by stylization is a promising yet challenging problem in computer vision and graphics research. In this work, we focus on stylizing photorealistic appearance renderings of a given surface mesh of arbitrary topology. Motivated by the recent surge of cross-modal supervision of the Contrastive Language-Image Pre-training (CLIP) model, we propose TANGO, which transfers the appearance style of a given 3D shape according to a text prompt in a photorealistic manner. Technically, we propose to disentangle the appearance style as the spatially varying bidirectional reflectance distribution function, the local geometric variation, and the lighting condition, which are jointly optimized, via supervision of the CLIP loss, by a spherical Gaussians based differentiable renderer. As such, TANGO enables photorealistic 3D style transfer by automatically predicting reflectance effects even for bare, low-quality meshes, without training on a task-specific dataset. Extensive experiments show that TANGO outperforms existing methods of text-driven 3D style transfer in terms of photorealistic quality, consistency of 3D geometry, and robustness when stylizing low-quality meshes. Our codes and results are available at our project webpage this https URL.

• Coordinates Are NOT Lonely -- Codebook Prior Helps Implicit Neural 3D Representations, NeurIPS2022 | [code]

  Implicit neural 3D representation has achieved impressive results in surface or scene reconstruction and novel view synthesis, which typically uses the coordinate-based multi-layer perceptrons (MLPs) to learn a continuous scene representation. However, existing approaches, such as Neural Radiance Field (NeRF) and its variants, usually require dense input views (i.e. 50-150) to obtain decent results. To relive the over-dependence on massive calibrated images and enrich the coordinate-based feature representation, we explore injecting the prior information into the coordinate-based network and introduce a novel coordinate-based model, CoCo-INR, for implicit neural 3D representation. The cores of our method are two attention modules: codebook attention and coordinate attention. The former extracts the useful prototypes containing rich geometry and appearance information from the prior codebook, and the latter propagates such prior information into each coordinate and enriches its feature representation for a scene or object surface. With the help of the prior information, our method can render 3D views with more photo-realistic appearance and geometries than the current methods using fewer calibrated images available. Experiments on various scene reconstruction datasets, including DTU and BlendedMVS, and the full 3D head reconstruction dataset, H3DS, demonstrate the robustness under fewer input views and fine detail-preserving capability of our proposed method.

• Parallel Inversion of Neural Radiance Fields for Robust Pose Estimation, ICRA2023 | [code]

  We present a parallelized optimization method based on fast Neural Radiance Fields (NeRF) for estimating 6-DoF target poses. Given a single observed RGB image of the target, we can predict the translation and rotation of the camera by minimizing the residual between pixels rendered from a fast NeRF model and pixels in the observed image. We integrate a momentum-based camera extrinsic optimization procedure into Instant Neural Graphics Primitives, a recent exceptionally fast NeRF implementation. By introducing parallel Monte Carlo sampling into the pose estimation task, our method overcomes local minima and improves efficiency in a more extensive search space. We also show the importance of adopting a more robust pixel-based loss function to reduce error. Experiments demonstrate that our method can achieve improved generalization and robustness on both synthetic and real-world benchmarks.

• Neural Contact Fields: Tracking Extrinsic Contact with Tactile Sensing | [code]

  We present Neural Contact Fields, a method that brings together neural fields and tactile sensing to address the problem of tracking extrinsic contact between object and environment. Knowing where the external contact occurs is a first step towards methods that can actively control it in facilitating downstream manipulation tasks. Prior work for localizing environmental contacts typically assume a contact type (e.g. point or line), does not capture contact/no-contact transitions, and only works with basic geometric-shaped objects. Neural Contact Fields are the first method that can track arbitrary multi-modal extrinsic contacts without making any assumptions about the contact type. Our key insight is to estimate the probability of contact for any 3D point in the latent space of object shapes, given vision-based tactile inputs that sense the local motion resulting from the external contact. In experiments, we find that Neural Contact Fields are able to localize multiple contact patches without making any assumptions about the geometry of the contact, and capture contact/no-contact transitions for known categories of objects with unseen shapes in unseen environment configurations. In addition to Neural Contact Fields, we also release our YCB-Extrinsic-Contact dataset of simulated extrinsic contact interactions to enable further research in this area. Project repository: this https URL

• S3-NeRF: Neural Reflectance Field from Shading and Shadow under a Single Viewpoint, NeurIPS2022 | [code]

  In this paper, we address the "dual problem" of multi-view scene reconstruction in which we utilize single-view images captured under different point lights to learn a neural scene representation. Different from existing single-view methods which can only recover a 2.5D scene representation (i.e., a normal / depth map for the visible surface), our method learns a neural reflectance field to represent the 3D geometry and BRDFs of a scene. Instead of relying on multi-view photo-consistency, our method exploits two information-rich monocular cues, namely shading and shadow, to infer scene geometry. Experiments on multiple challenging datasets show that our method is capable of recovering 3D geometry, including both visible and invisible parts, of a scene from single-view images. Thanks to the neural reflectance field representation, our method is robust to depth discontinuities. It supports applications like novel-view synthesis and relighting. Our code and model can be found at this https URL.

• Differentiable Physics Simulation of Dynamics-Augmented Neural Objects | [code]

  We present a differentiable pipeline for simulating the motion of objects that represent their geometry as a continuous density field parameterized as a deep network. This includes Neural Radiance Fields (NeRFs), and other related models. From the density field, we estimate the dynamical properties of the object, including its mass, center of mass, and inertia matrix. We then introduce a differentiable contact model based on the density field for computing normal and friction forces resulting from collisions. This allows a robot to autonomously build object models that are visually and dynamically accurate from still images and videos of objects in motion. The resulting Dynamics-Augmented Neural Objects (DANOs) are simulated with an existing differentiable simulation engine, Dojo, interacting with other standard simulation objects, such as spheres, planes, and robots specified as URDFs. A robot can use this simulation to optimize grasps and manipulation trajectories of neural objects, or to improve the neural object models through gradient-based real-to-simulation transfer. We demonstrate the pipeline to learn the coefficient of friction of a bar of soap from a real video of the soap sliding on a table. We also learn the coefficient of friction and mass of a Stanford bunny through interactions with a Panda robot arm from synthetic data, and we optimize trajectories in simulation for the Panda arm to push the bunny to a goal location.

Oct9 - Oct15, 2022

• LB-NERF: Light Bending Neural Radiance Fields for Transparent Medium, ICIP2022 | [code]

  Neural radiance fields (NeRFs) have been proposed as methods of novel view synthesis and have been used to address various problems because of its versatility. NeRF can represent colors and densities in 3D space using neural rendering assuming a straight light path. However, a medium with a different refractive index in the scene, such as a transparent medium, causes light refraction and breaks the assumption of the straight path of light. Therefore, the NeRFs cannot be learned consistently across multi-view images. To solve this problem, this study proposes a method to learn consistent radiance fields across multiple viewpoints by introducing the light refraction effect as an offset from the straight line originating from the camera center. The experimental results quantitatively and qualitatively verified that our method can interpolate viewpoints better than the conventional NeRF method when considering the refraction of transparent objects.

• IBL-NeRF: Image-Based Lighting Formulation of Neural Radiance Fields | [code]

  We propose IBL-NeRF, which decomposes the neural radiance fields (NeRF) of large-scale indoor scenes into intrinsic components. Previous approaches for the inverse rendering of NeRF transform the implicit volume to fit the rendering pipeline of explicit geometry, and approximate the views of segmented, isolated objects with environment lighting. In contrast, our inverse rendering extends the original NeRF formulation to capture the spatial variation of lighting within the scene volume, in addition to surface properties. Specifically, the scenes of diverse materials are decomposed into intrinsic components for image-based rendering, namely, albedo, roughness, surface normal, irradiance, and prefiltered radiance. All of the components are inferred as neural images from MLP, which can model large-scale general scenes. By adopting the image-based formulation of NeRF, our approach inherits superior visual quality and multi-view consistency for synthesized images. We demonstrate the performance on scenes with complex object layouts and light configurations, which could not be processed in any of the previous works.

• ExAug: Robot-Conditioned Navigation Policies via Geometric Experience Augmentation | [code]

  Machine learning techniques rely on large and diverse datasets for generalization. Computer vision, natural language processing, and other applications can often reuse public datasets to train many different models. However, due to differences in physical configurations, it is challenging to leverage public datasets for training robotic control policies on new robot platforms or for new tasks. In this work, we propose a novel framework, ExAug to augment the experiences of different robot platforms from multiple datasets in diverse environments. ExAug leverages a simple principle: by extracting 3D information in the form of a point cloud, we can create much more complex and structured augmentations, utilizing both generating synthetic images and geometric-aware penalization that would have been suitable in the same situation for a different robot, with different size, turning radius, and camera placement. The trained policy is evaluated on two new robot platforms with three different cameras in indoor and outdoor environments with obstacles.

• Lightweight Stepless Super-Resolution of Remote Sensing Images via Saliency-Aware Dynamic Routing Strategy | [code]

  Deep learning-based algorithms have greatly improved the performance of remote sensing image (RSI) super-resolution (SR). However, increasing network depth and parameters cause a huge burden of computing and storage. Directly reducing the depth or width of existing models results in a large performance drop. We observe that the SR difficulty of different regions in an RSI varies greatly, and existing methods use the same deep network to process all regions in an image, resulting in a waste of computing resources. In addition, existing SR methods generally predefine integer scale factors and cannot perform stepless SR, i.e., a single model can deal with any potential scale factor. Retraining the model on each scale factor wastes considerable computing resources and model storage space. To address the above problems, we propose a saliency-aware dynamic routing network (SalDRN) for lightweight and stepless SR of RSIs. First, we introduce visual saliency as an indicator of region-level SR difficulty and integrate a lightweight saliency detector into the SalDRN to capture pixel-level visual characteristics. Then, we devise a saliency-aware dynamic routing strategy that employs path selection switches to adaptively select feature extraction paths of appropriate depth according to the SR difficulty of sub-image patches. Finally, we propose a novel lightweight stepless upsampling module whose core is an implicit feature function for realizing mapping from low-resolution feature space to high-resolution feature space. Comprehensive experiments verify that the SalDRN can achieve a good trade-off between performance and complexity. The code is available at \url{this https URL}.

• NOCaL: Calibration-Free Semi-Supervised Learning of Odometry and Camera Intrinsics | [code]

  There are a multitude of emerging imaging technologies that could benefit robotics. However the need for bespoke models, calibration and low-level processing represents a key barrier to their adoption. In this work we present NOCaL, Neural odometry and Calibration using Light fields, a semi-supervised learning architecture capable of interpreting previously unseen cameras without calibration. NOCaL learns to estimate camera parameters, relative pose, and scene appearance. It employs a scene-rendering hypernetwork pretrained on a large number of existing cameras and scenes, and adapts to previously unseen cameras using a small supervised training set to enforce metric scale. We demonstrate NOCaL on rendered and captured imagery using conventional cameras, demonstrating calibration-free odometry and novel view synthesis. This work represents a key step toward automating the interpretation of general camera geometries and emerging imaging technologies.

• Multi-View Photometric Stereo Revisited, WACV2023 | [code]

  Multi-view photometric stereo (MVPS) is a preferred method for detailed and precise 3D acquisition of an object from images. Although popular methods for MVPS can provide outstanding results, they are often complex to execute and limited to isotropic material objects. To address such limitations, we present a simple, practical approach to MVPS, which works well for isotropic as well as other object material types such as anisotropic and glossy. The proposed approach in this paper exploits the benefit of uncertainty modeling in a deep neural network for a reliable fusion of photometric stereo (PS) and multi-view stereo (MVS) network predictions. Yet, contrary to the recently proposed state-of-the-art, we introduce neural volume rendering methodology for a trustworthy fusion of MVS and PS measurements. The advantage of introducing neural volume rendering is that it helps in the reliable modeling of objects with diverse material types, where existing MVS methods, PS methods, or both may fail. Furthermore, it allows us to work on neural 3D shape representation, which has recently shown outstanding results for many geometric processing tasks. Our suggested new loss function aims to fits the zero level set of the implicit neural function using the most certain MVS and PS network predictions coupled with weighted neural volume rendering cost. The proposed approach shows state-of-the-art results when tested extensively on several benchmark datasets.

• Controllable Style Transfer via Test-time Training of Implicit Neural Representation | [code]

  We propose a controllable style transfer framework based on Implicit Neural Representation that pixel-wisely controls the stylized output via test-time training. Unlike traditional image optimization methods that often suffer from unstable convergence and learning-based methods that require intensive training and have limited generalization ability, we present a model optimization framework that optimizes the neural networks during test-time with explicit loss functions for style transfer. After being test-time trained once, thanks to the flexibility of the INR-based model, our framework can precisely control the stylized images in a pixel-wise manner and freely adjust image resolution without further optimization or training. We demonstrate several applications.

• Scalable Neural Video Representations with Learnable Positional Features, NeurIPS2022 | [code]

  Succinct representation of complex signals using coordinate-based neural representations (CNRs) has seen great progress, and several recent efforts focus on extending them for handling videos. Here, the main challenge is how to (a) alleviate a compute-inefficiency in training CNRs to (b) achieve high-quality video encoding while (c) maintaining the parameter-efficiency. To meet all requirements (a), (b), and (c) simultaneously, we propose neural video representations with learnable positional features (NVP), a novel CNR by introducing "learnable positional features" that effectively amortize a video as latent codes. Specifically, we first present a CNR architecture based on designing 2D latent keyframes to learn the common video contents across each spatio-temporal axis, which dramatically improves all of those three requirements. Then, we propose to utilize existing powerful image and video codecs as a compute-/memory-efficient compression procedure of latent codes. We demonstrate the superiority of NVP on the popular UVG benchmark; compared with prior arts, NVP not only trains 2 times faster (less than 5 minutes) but also exceeds their encoding quality as 34.07→34.57 (measured with the PSNR metric), even using >8 times fewer parameters. We also show intriguing properties of NVP, e.g., video inpainting, video frame interpolation, etc.

• NeuralRoom: Geometry-Constrained Neural Implicit Surfaces for Indoor Scene Reconstruction | [code]

  We present a novel neural surface reconstruction method called NeuralRoom for reconstructing room-sized indoor scenes directly from a set of 2D images. Recently, implicit neural representations have become a promising way to reconstruct surfaces from multiview images due to their high-quality results and simplicity. However, implicit neural representations usually cannot reconstruct indoor scenes well because they suffer severe shape-radiance ambiguity. We assume that the indoor scene consists of texture-rich and flat texture-less regions. In texture-rich regions, the multiview stereo can obtain accurate results. In the flat area, normal estimation networks usually obtain a good normal estimation. Based on the above observations, we reduce the possible spatial variation range of implicit neural surfaces by reliable geometric priors to alleviate shape-radiance ambiguity. Specifically, we use multiview stereo results to limit the NeuralRoom optimization space and then use reliable geometric priors to guide NeuralRoom training. Then the NeuralRoom would produce a neural scene representation that can render an image consistent with the input training images. In addition, we propose a smoothing method called perturbation-residual restrictions to improve the accuracy and completeness of the flat region, which assumes that the sampling points in a local surface should have the same normal and similar distance to the observation center. Experiments on the ScanNet dataset show that our method can reconstruct the texture-less area of indoor scenes while maintaining the accuracy of detail. We also apply NeuralRoom to more advanced multiview reconstruction algorithms and significantly improve their reconstruction quality.

• CUF: Continuous Upsampling Filters | [code]

  Neural fields have rapidly been adopted for representing 3D signals, but their application to more classical 2D image-processing has been relatively limited. In this paper, we consider one of the most important operations in image processing: upsampling. In deep learning, learnable upsampling layers have extensively been used for single image super-resolution. We propose to parameterize upsampling kernels as neural fields. This parameterization leads to a compact architecture that obtains a 40-fold reduction in the number of parameters when compared with competing arbitrary-scale super-resolution architectures. When upsampling images of size 256x256 we show that our architecture is 2x-10x more efficient than competing arbitrary-scale super-resolution architectures, and more efficient than sub-pixel convolutions when instantiated to a single-scale model. In the general setting, these gains grow polynomially with the square of the target scale. We validate our method on standard benchmarks showing such efficiency gains can be achieved without sacrifices in super-resolution performance.

• GeoAug: Data Augmentation for Few-Shot NeRF with Geometry Constraints, ECCV2022 | [code]

  Neural Radiance Fields (NeRF) show remarkable ability to render novel views of a certain scene by learning an implicit volumetric representation with only posed RGB images. Despite its impressiveness and simplicity, NeRF usually converges to sub-optimal solutions with incorrect geometries given few training images. We hereby present GeoAug: a data augmentation method for NeRF, which enriches training data based on multi-view geometric constraint. GeoAug provides random artificial (novel pose, RGB image) pairs for training, where the RGB image is from a nearby training view. The rendering of a novel pose is warped to the nearby training view with depth map and relative pose to match the RGB image supervision. Our method reduces the risk of over-fitting by introducing more data during training, while also provides additional implicit supervision for depth maps. In experiments, our method significantly boosts the performance of neural radiance fields conditioned on few training views.

• Photo-realistic Neural Domain Randomization, ECCV2022 | [code]

  Synthetic data is a scalable alternative to manual supervision, but it requires overcoming the sim-to-real domain gap. This discrepancy between virtual and real worlds is addressed by two seemingly opposed approaches: improving the realism of simulation or foregoing realism entirely via domain randomization. In this paper, we show that the recent progress in neural rendering enables a new unified approach we call Photo-realistic Neural Domain Randomization (PNDR). We propose to learn a composition of neural networks that acts as a physics-based ray tracer generating high-quality renderings from scene geometry alone. Our approach is modular, composed of different neural networks for materials, lighting, and rendering, thus enabling randomization of different key image generation components in a differentiable pipeline. Once trained, our method can be combined with other methods and used to generate photo-realistic image augmentations online and significantly more efficiently than via traditional ray-tracing. We demonstrate the usefulness of PNDR through two downstream tasks: 6D object detection and monocular depth estimation. Our experiments show that training with PNDR enables generalization to novel scenes and significantly outperforms the state of the art in terms of real-world transfer.

• AniFaceGAN: Animatable 3D-Aware Face Image Generation for Video Avatars, NeurIPS2022 | [code]

  Although 2D generative models have made great progress in face image generation and animation, they often suffer from undesirable artifacts such as 3D inconsistency when rendering images from different camera viewpoints. This prevents them from synthesizing video animations indistinguishable from real ones. Recently, 3D-aware GANs extend 2D GANs for explicit disentanglement of camera pose by leveraging 3D scene representations. These methods can well preserve the 3D consistency of the generated images across different views, yet they cannot achieve fine-grained control over other attributes, among which facial expression control is arguably the most useful and desirable for face animation. In this paper, we propose an animatable 3D-aware GAN for multiview consistent face animation generation. The key idea is to decompose the 3D representation of the 3D-aware GAN into a template field and a deformation field, where the former represents different identities with a canonical expression, and the latter characterizes expression variations of each identity. To achieve meaningful control over facial expressions via deformation, we propose a 3D-level imitative learning scheme between the generator and a parametric 3D face model during adversarial training of the 3D-aware GAN. This helps our method achieve high-quality animatable face image generation with strong visual 3D consistency, even though trained with only unstructured 2D images. Extensive experiments demonstrate our superior performance over prior works. Project page: this https URL

• Reconstructing Personalized Semantic Facial NeRF Models From Monocular Video, SIGGRAPH-Asia2022 | [code]

  We present a novel semantic model for human head defined with neural radiance field. The 3D-consistent head model consist of a set of disentangled and interpretable bases, and can be driven by low-dimensional expression coefficients. Thanks to the powerful representation ability of neural radiance field, the constructed model can represent complex facial attributes including hair, wearings, which can not be represented by traditional mesh blendshape. To construct the personalized semantic facial model, we propose to define the bases as several multi-level voxel fields. With a short monocular RGB video as input, our method can construct the subject's semantic facial NeRF model with only ten to twenty minutes, and can render a photo-realistic human head image in tens of miliseconds with a given expression coefficient and view direction. With this novel representation, we apply it to many tasks like facial retargeting and expression editing. Experimental results demonstrate its strong representation ability and training/inference speed. Demo videos and released code are provided in our project page: this https URL

• LION: Latent Point Diffusion Models for 3D Shape Generation, NeurIPS2022 | [code]

  Denoising diffusion models (DDMs) have shown promising results in 3D point cloud synthesis. To advance 3D DDMs and make them useful for digital artists, we require (i) high generation quality, (ii) flexibility for manipulation and applications such as conditional synthesis and shape interpolation, and (iii) the ability to output smooth surfaces or meshes. To this end, we introduce the hierarchical Latent Point Diffusion Model (LION) for 3D shape generation. LION is set up as a variational autoencoder (VAE) with a hierarchical latent space that combines a global shape latent representation with a point-structured latent space. For generation, we train two hierarchical DDMs in these latent spaces. The hierarchical VAE approach boosts performance compared to DDMs that operate on point clouds directly, while the point-structured latents are still ideally suited for DDM-based modeling. Experimentally, LION achieves state-of-the-art generation performance on multiple ShapeNet benchmarks. Furthermore, our VAE framework allows us to easily use LION for different relevant tasks: LION excels at multimodal shape denoising and voxel-conditioned synthesis, and it can be adapted for text- and image-driven 3D generation. We also demonstrate shape autoencoding and latent shape interpolation, and we augment LION with modern surface reconstruction techniques to generate smooth 3D meshes. We hope that LION provides a powerful tool for artists working with 3D shapes due to its high-quality generation, flexibility, and surface reconstruction. Project page and code: this https URL.

• GraspNeRF: Multiview-based 6-DoF Grasp Detection for Transparent and Specular Objects Using Generalizable NeRF | [code]

  In this work, we tackle 6-DoF grasp detection for transparent and specular objects, which is an important yet challenging problem in vision-based robotic systems, due to the failure of depth cameras in sensing their geometry. We, for the first time, propose a multiview RGB-based 6-DoF grasp detection network, GraspNeRF, that leverages the generalizable neural radiance field (NeRF) to achieve material-agnostic object grasping in clutter. Compared to the existing NeRF-based 3-DoF grasp detection methods that rely on densely captured input images and time-consuming per-scene optimization, our system can perform zero-shot NeRF construction with sparse RGB inputs and reliably detect 6-DoF grasps, both in real-time. The proposed framework jointly learns generalizable NeRF and grasp detection in an end-to-end manner, optimizing the scene representation construction for the grasping. For training data, we generate a large-scale photorealistic domain-randomized synthetic dataset of grasping in cluttered tabletop scenes that enables direct transfer to the real world. Our extensive experiments in synthetic and real-world environments demonstrate that our method significantly outperforms all the baselines in all the experiments while remaining in real-time.

• X-NeRF: Explicit Neural Radiance Field for Multi-Scene 360∘ Insufficient RGB-D Views, WACV2023 | [code]

  Neural Radiance Fields (NeRFs), despite their outstanding performance on novel view synthesis, often need dense input views. Many papers train one model for each scene respectively and few of them explore incorporating multi-modal data into this problem. In this paper, we focus on a rarely discussed but important setting: can we train one model that can represent multiple scenes, with 360∘ insufficient views and RGB-D images? We refer insufficient views to few extremely sparse and almost non-overlapping views. To deal with it, X-NeRF, a fully explicit approach which learns a general scene completion process instead of a coordinate-based mapping, is proposed. Given a few insufficient RGB-D input views, X-NeRF first transforms them to a sparse point cloud tensor and then applies a 3D sparse generative Convolutional Neural Network (CNN) to complete it to an explicit radiance field whose volumetric rendering can be conducted fast without running networks during inference. To avoid overfitting, besides common rendering loss, we apply perceptual loss as well as view augmentation through random rotation on point clouds. The proposed methodology significantly out-performs previous implicit methods in our setting, indicating the great potential of proposed problem and approach. Codes and data are available at this https URL.

• Multi-Object Navigation with dynamically learned neural implicit representations | [code]

  Understanding and mapping a new environment are core abilities of any autonomously navigating agent. While classical robotics usually estimates maps in a stand-alone manner with SLAM variants, which maintain a topological or metric representation, end-to-end learning of navigation keeps some form of memory in a neural network. Networks are typically imbued with inductive biases, which can range from vectorial representations to birds-eye metric tensors or topological structures. In this work, we propose to structure neural networks with two neural implicit representations, which are learned dynamically during each episode and map the content of the scene: (i) the Semantic Finder predicts the position of a previously seen queried object; (ii) the Occupancy and Exploration Implicit Representation encapsulates information about explored area and obstacles, and is queried with a novel global read mechanism which directly maps from function space to a usable embedding space. Both representations are leveraged by an agent trained with Reinforcement Learning (RL) and learned online during each episode. We evaluate the agent on Multi-Object Navigation and show the high impact of using neural implicit representations as a memory source.

• CLIP-Fields: Weakly Supervised Semantic Fields for Robotic Memory | [code]

  We propose CLIP-Fields, an implicit scene model that can be trained with no direct human supervision. This model learns a mapping from spatial locations to semantic embedding vectors. The mapping can then be used for a variety of tasks, such as segmentation, instance identification, semantic search over space, and view localization. Most importantly, the mapping can be trained with supervision coming only from web-image and web-text trained models such as CLIP, Detic, and Sentence-BERT. When compared to baselines like Mask-RCNN, our method outperforms on few-shot instance identification or semantic segmentation on the HM3D dataset with only a fraction of the examples. Finally, we show that using CLIP-Fields as a scene memory, robots can perform semantic navigation in real-world environments. Our code and demonstrations are available here: https://mahis.life/clip-fields/

• Neural Shape Deformation Priors, NeurIPS2022 | [code]

  We present Neural Shape Deformation Priors, a novel method for shape manipulation that predicts mesh deformations of non-rigid objects from user-provided handle movements. State-of-the-art methods cast this problem as an optimization task, where the input source mesh is iteratively deformed to minimize an objective function according to hand-crafted regularizers such as ARAP. In this work, we learn the deformation behavior based on the underlying geometric properties of a shape, while leveraging a large-scale dataset containing a diverse set of non-rigid deformations. Specifically, given a source mesh and desired target locations of handles that describe the partial surface deformation, we predict a continuous deformation field that is defined in 3D space to describe the space deformation. To this end, we introduce transformer-based deformation networks that represent a shape deformation as a composition of local surface deformations. It learns a set of local latent codes anchored in 3D space, from which we can learn a set of continuous deformation functions for local surfaces. Our method can be applied to challenging deformations and generalizes well to unseen deformations. We validate our approach in experiments using the DeformingThing4D dataset, and compare to both classic optimization-based and recent neural network-based methods.

• Controllable Radiance Fields for Dynamic Face Synthesis, 3DV2022 | [code]

  Recent work on 3D-aware image synthesis has achieved compelling results using advances in neural rendering. However, 3D-aware synthesis of face dynamics hasn't received much attention. Here, we study how to explicitly control generative model synthesis of face dynamics exhibiting non-rigid motion (e.g., facial expression change), while simultaneously ensuring 3D-awareness. For this we propose a Controllable Radiance Field (CoRF): 1) Motion control is achieved by embedding motion features within the layered latent motion space of a style-based generator; 2) To ensure consistency of background, motion features and subject-specific attributes such as lighting, texture, shapes, albedo, and identity, a face parsing net, a head regressor and an identity encoder are incorporated. On head image/video data we show that CoRFs are 3D-aware while enabling editing of identity, viewing directions, and motion.

• Continuous conditional video synthesis by neural processes | [code]

  We propose a unified model for multiple conditional video synthesis tasks, including video prediction and video frame interpolation. We show that conditional video synthesis can be formulated as a neural process, which maps input spatio-temporal coordinates to target pixel values given context spatio-temporal coordinates and pixels values. Specifically, we feed an implicit neural representations of coordinates into a Transformer-based non-autoregressive conditional video synthesis model. Our task-specific models outperform previous work for video interpolation on multiple datasets and reach a competitive performance with the state-of-the-art models for video prediction. Importantly, the model is able to interpolate or predict with an arbitrary high frame rate, i.e., continuous synthesis. Our source code is available at this https URL.

• SiNeRF: Sinusoidal Neural Radiance Fields for Joint Pose Estimation and Scene Reconstruction, BMVC2022 | [code]

  NeRFmm is the Neural Radiance Fields (NeRF) that deal with Joint Optimization tasks, i.e., reconstructing real-world scenes and registering camera parameters simultaneously. Despite NeRFmm producing precise scene synthesis and pose estimations, it still struggles to outperform the full-annotated baseline on challenging scenes. In this work, we identify that there exists a systematic sub-optimality in joint optimization and further identify multiple potential sources for it. To diminish the impacts of potential sources, we propose Sinusoidal Neural Radiance Fields (SiNeRF) that leverage sinusoidal activations for radiance mapping and a novel Mixed Region Sampling (MRS) for selecting ray batch efficiently. Quantitative and qualitative results show that compared to NeRFmm, SiNeRF achieves comprehensive significant improvements in image synthesis quality and pose estimation accuracy. Codes are available at this https URL.

• NerfAcc: A General NeRF Acceleration Toolbox | [code]

  We propose NerfAcc, a toolbox for efficient volumetric rendering of radiance fields. We build on the techniques proposed in Instant-NGP, and extend these techniques to not only support bounded static scenes, but also for dynamic scenes and unbounded scenes. NerfAcc comes with a user-friendly Python API, and is ready for plug-and-play acceleration of most NeRFs. Various examples are provided to show how to use this toolbox. Code can be found here: this https URL.

• Self-Supervised 3D Human Pose Estimation in Static Video Via Neural Rendering | [code]

  Inferring 3D human pose from 2D images is a challenging and long-standing problem in the field of computer vision with many applications including motion capture, virtual reality, surveillance or gait analysis for sports and medicine. We present preliminary results for a method to estimate 3D pose from 2D video containing a single person and a static background without the need for any manual landmark annotations. We achieve this by formulating a simple yet effective self-supervision task: our model is required to reconstruct a random frame of a video given a frame from another timepoint and a rendered image of a transformed human shape template. Crucially for optimisation, our ray casting based rendering pipeline is fully differentiable, enabling end to end training solely based on the reconstruction task.

• MVSPlenOctree: Fast and Generic Reconstruction of Radiance Fields in PlenOctree from Multi-view Stereo, ACMMM2022 | [code]

  We present MVSPlenOctree, a novel approach that can efficiently reconstruct radiance fields for view synthesis. Unlike previous scene-specific radiance fields reconstruction methods, we present a generic pipeline that can efficiently reconstruct 360-degree-renderable radiance fields via multi-view stereo (MVS) inference from tens of sparse-spread out images. Our approach leverages variance-based statistic features for MVS inference, and combines this with image based rendering and volume rendering for radiance field reconstruction. We first train a MVS Machine for reasoning scene's density and appearance. Then, based on the spatial hierarchy of the PlenOctree and coarse-to-fine dense sampling mechanism, we design a robust and efficient sampling strategy for PlenOctree reconstruction, which handles occlusion robustly. A 360-degree-renderable radiance fields can be reconstructed in PlenOctree from MVS Machine in an efficient single forward pass. We trained our method on real-world DTU, LLFF datasets, and synthetic datasets. We validate its generalizability by evaluating on the test set of DTU dataset which are unseen in training. In summary, our radiance field reconstruction method is both efficient and generic, a coarse 360-degree-renderable radiance field can be reconstructed in seconds and a dense one within minutes. Please visit the project page for more details: https://derry-xing.github.io/projects/MVSPlenOctree.

• ParseMVS: Learning Primitive-aware Surface Representations for Sparse Multi-view Stereopsis, ACMMM2022 | [code]

  Multi-view stereopsis (MVS) recovers 3D surfaces by finding dense photo-consistent correspondences from densely sampled images. In this paper, we tackle the challenging MVS task from sparsely sampled views (up to an order of magnitude fewer images), which is more practical and cost-efficient in applications. The major challenge comes from the significant correspondence ambiguity introduced by the severe occlusions and the highly skewed patches. On the other hand, such ambiguity can be resolved by incorporating geometric cues from the global structure. In light of this, we propose ParseMVS, boosting sparse MVS by learning the P rimitive-A waR e S urface rE presentation. In particular, on top of being aware of global structure, our novel representation further allows for the preservation of fine details including geometry, texture, and visibility. More specifically, the whole scene is parsed into multiple geometric primitives. On each of them, the geometry is defined as the displacement along the primitives' normal directions, together with the texture and visibility along each view direction. An unsupervised neural network is trained to learn these factors by progressively increasing the photo-consistency and render-consistency among all input images. Since the surface properties are changed locally in the 2D space of each primitive, ParseMVS can preserve global primitive structures while optimizing local details, handling the 'incompleteness' and the 'inaccuracy' problems. We experimentally demonstrate that ParseMVS constantly outperforms the state-of-the-art surface reconstruction method in both completeness and the overall score under varying sampling sparsity, especially under the extreme sparse-MVS settings. Beyond that, ParseMVS also shows great potential in compression, robustness, and efficiency.

• Self-Supervised Multi-view Stereo via Adjacent Geometry Guided Volume Completion, ACMMM2022 | [code]

  Existing self-supervised multi-view stereo (MVS) approaches largely rely on photometric consistency for geometry inference, and hence suffer from low-texture or non-Lambertian appearances. In this paper, we observe that adjacent geometry shares certain commonality that can help to infer the correct geometry of the challenging or low-confident regions. Yet exploiting such property in a non-supervised MVS approach remains challenging for the lacking of training data and necessity of ensuring consistency between views. To address the issues, we propose a novel geometry inference training scheme by selectively masking regions with rich textures, where geometry can be well recovered and used for supervisory signal, and then lead a deliberately designed cost volume completion network to learn how to recover geometry of the masked regions. During inference, we then mask the low-confident regions instead and use the cost volume completion network for geometry correction. To deal with the different depth hypotheses of the cost volume pyramid, we design a three-branch volume inference structure for the completion network. Further, by considering plane as a special geometry, we first identify planar regions from pseudo labels and then correct the low-confident pixels by high-confident labels through plane normal consistency. Extensive experiments on DTU and Tanks & Temples demonstrate the effectiveness of the proposed framework and the state-of-the-art performance.

• Geometric Warping Error Aware CNN for DIBR Oriented View Synthesis, ACMMM2022 | [code]

  Depth Image based Rendering (DIBR) oriented view synthesis is an important virtual view generation technique. It warps the reference view images to the target viewpoint based on their depth maps, without requiring many available viewpoints. However, in the 3D warping process, pixels are warped to fractional pixel locations and then rounded (or interpolated) to integer pixels, resulting in geometric warping error and reducing the image quality. This resembles, to some extent, the image super-resolution problem, but with unfixed fractional pixel locations. To address this problem, we propose a geometric warping error aware CNN (GWEA) framework to enhance the DIBR oriented view synthesis. First, a deformable convolution based geometric warping error aware alignment (GWEA-DCA) module is developed, by taking advantage of the geometric warping error preserved in the DIBR module. The offset learned in the deformable convolution can account for the geometric warping error to facilitate the mapping from the fractional pixels to integer pixels. Moreover, in view that the pixels in the warped images are of different qualities due to the different strengths of warping errors, an attention enhanced view blending (GWEA-AttVB) module is further developed to adaptively fuse the pixels from different warped images. Finally, a partial convolution based hole filling and refinement module fills the remaining holes and improves the quality of the overall image. Experiments show that our model can synthesize higher-quality images than the existing methods, and ablation study is also conducted, validating the effectiveness of each proposed module.

• ReFu: Refine and Fuse the Unobserved View for Detail-Preserving Single-Image 3D Human Reconstruction | [code]

  Single-image 3D human reconstruction aims to reconstruct the 3D textured surface of the human body given a single image. While implicit function-based methods recently achieved reasonable reconstruction performance, they still bear limitations showing degraded quality in both surface geometry and texture from an unobserved view. In response, to generate a realistic textured surface, we propose ReFu, a coarse-to-fine approach that refines the projected backside view image and fuses the refined image to predict the final human body. To suppress the diffused occupancy that causes noise in projection images and reconstructed meshes, we propose to train occupancy probability by simultaneously utilizing 2D and 3D supervisions with occupancy-based volume rendering. We also introduce a refinement architecture that generates detail-preserving backside-view images with front-to-back warping. Extensive experiments demonstrate that our method achieves state-of-the-art performance in 3D human reconstruction from a single image, showing enhanced geometry and texture quality from an unobserved view.

• NeRF2Real: Sim2real Transfer of Vision-guided Bipedal Motion Skills using Neural Radiance Fields | [code]

  We present a system for applying sim2real approaches to "in the wild" scenes with realistic visuals, and to policies which rely on active perception using RGB cameras. Given a short video of a static scene collected using a generic phone, we learn the scene's contact geometry and a function for novel view synthesis using a Neural Radiance Field (NeRF). We augment the NeRF rendering of the static scene by overlaying the rendering of other dynamic objects (e.g. the robot's own body, a ball). A simulation is then created using the rendering engine in a physics simulator which computes contact dynamics from the static scene geometry (estimated from the NeRF volume density) and the dynamic objects' geometry and physical properties (assumed known). We demonstrate that we can use this simulation to learn vision-based whole body navigation and ball pushing policies for a 20 degrees of freedom humanoid robot with an actuated head-mounted RGB camera, and we successfully transfer these policies to a real robot. Project video is available at this https URL

• Uncertainty-Aware Semi-Supervised Learning of 3D Face Rigging from Single Image, ACMMM2022 | [code]

  We present a method to rig 3D faces via Action Units (AUs), viewpoint and light direction, from single input image. Existing 3D methods for face synthesis and animation rely heavily on 3D morphable model (3DMM), which was built on 3D data and cannot provide intuitive expression parameters, while AU-driven 2D methods cannot handle head pose and lighting effect. We bridge the gap by integrating a recent 3D reconstruction method with 2D AU-driven method in a semi-supervised fashion. Built upon the auto-encoding 3D face reconstruction model that decouples depth, albedo, viewpoint and light without any supervision, we further decouple expression from identity for depth and albedo with a novel conditional feature translation module and pretrained critics for AU intensity estimation and image classification. Novel objective functions are designed using unlabeled in-the-wild images and in-door images with AU labels. We also leverage uncertainty losses to model the probably changing AU region of images as input noise for synthesis, and model the noisy AU intensity labels for intensity estimation of the AU critic. Experiments with face editing and animation on four datasets show that, compared with six state-of-the-art methods, our proposed method is superior and effective on expression consistency, identity similarity and pose similarity.

• Robustifying the Multi-Scale Representation of Neural Radiance Fields, BMVC2022 | [code]

  Neural Radiance Fields (NeRF) recently emerged as a new paradigm for object representation from multi-view (MV) images. Yet, it cannot handle multi-scale (MS) images and camera pose estimation errors, which generally is the case with multi-view images captured from a day-to-day commodity camera. Although recently proposed Mip-NeRF could handle multi-scale imaging problems with NeRF, it cannot handle camera pose estimation error. On the other hand, the newly proposed BARF can solve the camera pose problem with NeRF but fails if the images are multi-scale in nature. This paper presents a robust multi-scale neural radiance fields representation approach to simultaneously overcome both real-world imaging issues. Our method handles multi-scale imaging effects and camera-pose estimation problems with NeRF-inspired approaches by leveraging the fundamentals of scene rigidity. To reduce unpleasant aliasing artifacts due to multi-scale images in the ray space, we leverage Mip-NeRF multi-scale representation. For joint estimation of robust camera pose, we propose graph-neural network-based multiple motion averaging in the neural volume rendering framework. We demonstrate, with examples, that for an accurate neural representation of an object from day-to-day acquired multi-view images, it is crucial to have precise camera-pose estimates. Without considering robustness measures in the camera pose estimation, modeling for multi-scale aliasing artifacts via conical frustum can be counterproductive. We present extensive experiments on the benchmark datasets to demonstrate that our approach provides better results than the recent NeRF-inspired approaches for such realistic settings.

• Estimating Neural Reflectance Field from Radiance Field using Tree Structures | [code]

  We present a new method for estimating the Neural Reflectance Field (NReF) of an object from a set of posed multi-view images under unknown lighting. NReF represents 3D geometry and appearance of objects in a disentangled manner, and are hard to be estimated from images only. Our method solves this problem by exploiting the Neural Radiance Field (NeRF) as a proxy representation, from which we perform further decomposition. A high-quality NeRF decomposition relies on good geometry information extraction as well as good prior terms to properly resolve ambiguities between different components. To extract high-quality geometry information from radiance fields, we re-design a new ray-casting based method for surface point extraction. To efficiently compute and apply prior terms, we convert different prior terms into different type of filter operations on the surface extracted from radiance field. We then employ two type of auxiliary data structures, namely Gaussian KD-tree and octree, to support fast querying of surface points and efficient computation of surface filters during training. Based on this, we design a multi-stage decomposition optimization pipeline for estimating neural reflectance field from neural radiance fields. Extensive experiments show our method outperforms other state-of-the-art methods on different data, and enable high-quality free-view relighting as well as material editing tasks.

• Towards Efficient Neural Scene Graphs by Learning Consistency Fields, BMVC2022 | [code]

  Neural Radiance Fields (NeRF) achieves photo-realistic image rendering from novel views, and the Neural Scene Graphs (NSG) \cite{ost2021neural} extends it to dynamic scenes (video) with multiple objects. Nevertheless, computationally heavy ray marching for every image frame becomes a huge burden. In this paper, taking advantage of significant redundancy across adjacent frames in videos, we propose a feature-reusing framework. From the first try of naively reusing the NSG features, however, we learn that it is crucial to disentangle object-intrinsic properties consistent across frames from transient ones. Our proposed method, \textit{Consistency-Field-based NSG (CF-NSG)}, reformulates neural radiance fields to additionally consider \textit{consistency fields}. With disentangled representations, CF-NSG takes full advantage of the feature-reusing scheme and performs an extended degree of scene manipulation in a more controllable manner. We empirically verify that CF-NSG greatly improves the inference efficiency by using 85% less queries than NSG without notable degradation in rendering quality. Code will be available at: this https URL

Oct2 - Oct8, 2022

• ViewFool: Evaluating the Robustness of Visual Recognition to Adversarial Viewpoints, NeurIPS2022 | [code]

  Recent studies have demonstrated that visual recognition models lack robustness to distribution shift. However, current work mainly considers model robustness to 2D image transformations, leaving viewpoint changes in the 3D world less explored. In general, viewpoint changes are prevalent in various real-world applications (e.g., autonomous driving), making it imperative to evaluate viewpoint robustness. In this paper, we propose a novel method called ViewFool to find adversarial viewpoints that mislead visual recognition models. By encoding real-world objects as neural radiance fields (NeRF), ViewFool characterizes a distribution of diverse adversarial viewpoints under an entropic regularizer, which helps to handle the fluctuations of the real camera pose and mitigate the reality gap between the real objects and their neural representations. Experiments validate that the common image classifiers are extremely vulnerable to the generated adversarial viewpoints, which also exhibit high cross-model transferability. Based on ViewFool, we introduce ImageNet-V, a new out-of-distribution dataset for benchmarking viewpoint robustness of image classifiers. Evaluation results on 40 classifiers with diverse architectures, objective functions, and data augmentations reveal a significant drop in model performance when tested on ImageNet-V, which provides a possibility to leverage ViewFool as an effective data augmentation strategy to improve viewpoint robustness.

• Novel View Synthesis for Surgical Recording | [code]

  Recording surgery in operating rooms is one of the essential tasks for education and evaluation of medical treatment. However, recording the fields which depict the surgery is difficult because the targets are heavily occluded during surgery by the heads or hands of doctors or nurses. We use a recording system which multiple cameras embedded in the surgical lamp, assuming that at least one camera is recording the target without occlusion. In this paper, we propose Conditional-BARF (C-BARF) to generate occlusion-free images by synthesizing novel view images from the camera, aiming to generate videos with smooth camera pose transitions. To the best of our knowledge, this is the first work to tackle the problem of synthesizing a novel view image from multiple images for the surgery scene. We conduct experiments using an original dataset of three different types of surgeries. Our experiments show that we can successfully synthesize novel views from the images recorded by the multiple cameras embedded in the surgical lamp.

• A Keypoint Based Enhancement Method for Audio Driven Free View Talking Head Synthesis | [code]

  Audio driven talking head synthesis is a challenging task that attracts increasing attention in recent years. Although existing methods based on 2D landmarks or 3D face models can synthesize accurate lip synchronization and rhythmic head pose for arbitrary identity, they still have limitations, such as the cut feeling in the mouth mapping and the lack of skin highlights. The morphed region is blurry compared to the surrounding face. A Keypoint Based Enhancement (KPBE) method is proposed for audio driven free view talking head synthesis to improve the naturalness of the generated video. Firstly, existing methods were used as the backend to synthesize intermediate results. Then we used keypoint decomposition to extract video synthesis controlling parameters from the backend output and the source image. After that, the controlling parameters were composited to the source keypoints and the driving keypoints. A motion field based method was used to generate the final image from the keypoint representation. With keypoint representation, we overcame the cut feeling in the mouth mapping and the lack of skin highlights. Experiments show that our proposed enhancement method improved the quality of talking-head videos in terms of mean opinion score.

• A Simple Plugin for Transforming Images to Arbitrary Scales | [code]

  Existing models on super-resolution often specialized for one scale, fundamentally limiting their use in practical scenarios. In this paper, we aim to develop a general plugin that can be inserted into existing super-resolution models, conveniently augmenting their ability towards Arbitrary Resolution Image Scaling, thus termed ARIS. We make the following contributions: (i) we propose a transformer-based plugin module, which uses spatial coordinates as query, iteratively attend the low-resolution image feature through cross-attention, and output visual feature for the queried spatial location, resembling an implicit representation for images; (ii) we introduce a novel self-supervised training scheme, that exploits consistency constraints to effectively augment the model's ability for upsampling images towards unseen scales, i.e. ground-truth high-resolution images are not available; (iii) without loss of generality, we inject the proposed ARIS plugin module into several existing models, namely, IPT, SwinIR, and HAT, showing that the resulting models can not only maintain their original performance on fixed scale factor but also extrapolate to unseen scales, substantially outperforming existing any-scale super-resolution models on standard benchmarks, e.g. Urban100, DIV2K, etc.

• Feature-Realistic Neural Fusion for Real-Time, Open Set Scene Understanding | [code]

  General scene understanding for robotics requires flexible semantic representation, so that novel objects and structures which may not have been known at training time can be identified, segmented and grouped. We present an algorithm which fuses general learned features from a standard pre-trained network into a highly efficient 3D geometric neural field representation during real-time SLAM. The fused 3D feature maps inherit the coherence of the neural field's geometry representation. This means that tiny amounts of human labelling interacting at runtime enable objects or even parts of objects to be robustly and accurately segmented in an open set manner.

• XDGAN: Multi-Modal 3D Shape Generation in 2D Space | [code]

  Generative models for 2D images has recently seen tremendous progress in quality, resolution and speed as a result of the efficiency of 2D convolutional architectures. However it is difficult to extend this progress into the 3D domain since most current 3D representations rely on custom network components. This paper addresses a central question: Is it possible to directly leverage 2D image generative models to generate 3D shapes instead? To answer this, we propose XDGAN, an effective and fast method for applying 2D image GAN architectures to the generation of 3D object geometry combined with additional surface attributes, like color textures and normals. Specifically, we propose a novel method to convert 3D shapes into compact 1-channel geometry images and leverage StyleGAN3 and image-to-image translation networks to generate 3D objects in 2D space. The generated geometry images are quick to convert to 3D meshes, enabling real-time 3D object synthesis, visualization and interactive editing. Moreover, the use of standard 2D architectures can help bring more 2D advances into the 3D realm. We show both quantitatively and qualitatively that our method is highly effective at various tasks such as 3D shape generation, single view reconstruction and shape manipulation, while being significantly faster and more flexible compared to recent 3D generative models.

• A Real2Sim2Real Method for Robust Object Grasping with Neural Surface Reconstruction | [code]

  Recent 3D-based manipulation methods either directly predict the grasp pose using 3D neural networks, or solve the grasp pose using similar objects retrieved from shape databases. However, the former faces generalizability challenges when testing with new robot arms or unseen objects; and the latter assumes that similar objects exist in the databases. We hypothesize that recent 3D modeling methods provides a path towards building digital replica of the evaluation scene that affords physical simulation and supports robust manipulation algorithm learning. We propose to reconstruct high-quality meshes from real-world point clouds using state-of-the-art neural surface reconstruction method (the Real2Sim step). Because most simulators take meshes for fast simulation, the reconstructed meshes enable grasp pose labels generation without human efforts. The generated labels can train grasp network that performs robustly in the real evaluation scene (the Sim2Real step). In synthetic and real experiments, we show that the Real2Sim2Real pipeline performs better than baseline grasp networks trained with a large dataset and a grasp sampling method with retrieval-based reconstruction. The benefit of the Real2Sim2Real pipeline comes from 1) decoupling scene modeling and grasp sampling into sub-problems, and 2) both sub-problems can be solved with sufficiently high quality using recent 3D learning algorithms and mesh-based physical simulation techniques.

• Feature-Realistic Neural Fusion for Real-Time, Open Set Scene Understanding | [code]

  General scene understanding for robotics requires flexible semantic representation, so that novel objects and structures which may not have been known at training time can be identified, segmented and grouped. We present an algorithm which fuses general learned features from a standard pre-trained network into a highly efficient 3D geometric neural field representation during real-time SLAM. The fused 3D feature maps inherit the coherence of the neural field's geometry representation. This means that tiny amounts of human labelling interacting at runtime enable objects or even parts of objects to be robustly and accurately segmented in an open set manner.

• Neural Matching Fields: Implicit Representation of Matching Fields for Visual Correspondence, NeurIPS2022 | [code]

  Existing pipelines of semantic correspondence commonly include extracting high-level semantic features for the invariance against intra-class variations and background clutters. This architecture, however, inevitably results in a low-resolution matching field that additionally requires an ad-hoc interpolation process as a post-processing for converting it into a high-resolution one, certainly limiting the overall performance of matching results. To overcome this, inspired by recent success of implicit neural representation, we present a novel method for semantic correspondence, called Neural Matching Field (NeMF). However, complicacy and high-dimensionality of a 4D matching field are the major hindrances, which we propose a cost embedding network to process a coarse cost volume to use as a guidance for establishing high-precision matching field through the following fully-connected network. Nevertheless, learning a high-dimensional matching field remains challenging mainly due to computational complexity, since a naive exhaustive inference would require querying from all pixels in the 4D space to infer pixel-wise correspondences. To overcome this, we propose adequate training and inference procedures, which in the training phase, we randomly sample matching candidates and in the inference phase, we iteratively performs PatchMatch-based inference and coordinate optimization at test time. With these combined, competitive results are attained on several standard benchmarks for semantic correspondence. Code and pre-trained weights are available at this https URL.

• IR-MCL: Implicit Representation-Based Online Global Localization | [code]

  Determining the state of a mobile robot is an essential building block of robot navigation systems. In this paper, we address the problem of estimating the robots pose in an indoor environment using 2D LiDAR data and investigate how modern environment models can improve gold standard Monte-Carlo localization (MCL) systems. We propose a neural occupancy field (NOF) to implicitly represent the scene using a neural network. With the pretrained network, we can synthesize 2D LiDAR scans for an arbitrary robot pose through volume rendering. Based on the implicit representation, we can obtain the similarity between a synthesized and actual scan as an observation model and integrate it into an MCL system to perform accurate localization. We evaluate our approach on five sequences of a self-recorded dataset and three publicly available datasets. We show that we can accurately and efficiently localize a robot using our approach surpassing the localization performance of state-of-the-art methods. The experiments suggest that the presented implicit representation is able to predict more accurate 2D LiDAR scans leading to an improved observation model for our particle filter-based localization. The code of our approach is released at: this https URL.

• SelfNeRF: Fast Training NeRF for Human from Monocular Self-rotating Video | [code]

  In this paper, we propose SelfNeRF, an efficient neural radiance field based novel view synthesis method for human performance. Given monocular self-rotating videos of human performers, SelfNeRF can train from scratch and achieve high-fidelity results in about twenty minutes. Some recent works have utilized the neural radiance field for dynamic human reconstruction. However, most of these methods need multi-view inputs and require hours of training, making it still difficult for practical use. To address this challenging problem, we introduce a surface-relative representation based on multi-resolution hash encoding that can greatly improve the training speed and aggregate inter-frame information. Extensive experimental results on several different datasets demonstrate the effectiveness and efficiency of SelfNeRF to challenging monocular videos.

• Capturing and Animation of Body and Clothing from Monocular Video | [code]

  While recent work has shown progress on extracting clothed 3D human avatars from a single image, video, or a set of 3D scans, several limitations remain. Most methods use a holistic representation to jointly model the body and clothing, which means that the clothing and body cannot be separated for applications like virtual try-on. Other methods separately model the body and clothing, but they require training from a large set of 3D clothed human meshes obtained from 3D/4D scanners or physics simulations. Our insight is that the body and clothing have different modeling requirements. While the body is well represented by a mesh-based parametric 3D model, implicit representations and neural radiance fields are better suited to capturing the large variety in shape and appearance present in clothing. Building on this insight, we propose SCARF (Segmented Clothed Avatar Radiance Field), a hybrid model combining a mesh-based body with a neural radiance field. Integrating the mesh into the volumetric rendering in combination with a differentiable rasterizer enables us to optimize SCARF directly from monocular videos, without any 3D supervision. The hybrid modeling enables SCARF to (i) animate the clothed body avatar by changing body poses (including hand articulation and facial expressions), (ii) synthesize novel views of the avatar, and (iii) transfer clothing between avatars in virtual try-on applications. We demonstrate that SCARF reconstructs clothing with higher visual quality than existing methods, that the clothing deforms with changing body pose and body shape, and that clothing can be successfully transferred between avatars of different subjects. The code and models are available at this https URL.

• Learning Perception-Aware Agile Flight in Cluttered Environments | [code]

  Recently, neural control policies have outperformed existing model-based planning-and-control methods for autonomously navigating quadrotors through cluttered environments in minimum time. However, they are not perception aware, a crucial requirement in vision-based navigation due to the camera's limited field of view and the underactuated nature of a quadrotor. We propose a method to learn neural network policies that achieve perception-aware, minimum-time flight in cluttered environments. Our method combines imitation learning and reinforcement learning (RL) by leveraging a privileged learning-by-cheating framework. Using RL, we first train a perception-aware teacher policy with full-state information to fly in minimum time through cluttered environments. Then, we use imitation learning to distill its knowledge into a vision-based student policy that only perceives the environment via a camera. Our approach tightly couples perception and control, showing a significant advantage in computation speed (10x faster) and success rate. We demonstrate the closed-loop control performance using a physical quadrotor and hardware-in-the-loop simulation at speeds up to 50km/h.

• Dfferentiable Raycasting for Self-supervised Occupancy Forecasting, ECCV2022 | [code]

  Motion planning for safe autonomous driving requires learning how the environment around an ego-vehicle evolves with time. Ego-centric perception of driveable regions in a scene not only changes with the motion of actors in the environment, but also with the movement of the ego-vehicle itself. Self-supervised representations proposed for large-scale planning, such as ego-centric freespace, confound these two motions, making the representation difficult to use for downstream motion planners. In this paper, we use geometric occupancy as a natural alternative to view-dependent representations such as freespace. Occupancy maps naturally disentangle the motion of the environment from the motion of the ego-vehicle. However, one cannot directly observe the full 3D occupancy of a scene (due to occlusion), making it difficult to use as a signal for learning. Our key insight is to use differentiable raycasting to "render" future occupancy predictions into future LiDAR sweep predictions, which can be compared with ground-truth sweeps for self-supervised learning. The use of differentiable raycasting allows occupancy to emerge as an internal representation within the forecasting network. In the absence of groundtruth occupancy, we quantitatively evaluate the forecasting of raycasted LiDAR sweeps and show improvements of upto 15 F1 points. For downstream motion planners, where emergent occupancy can be directly used to guide non-driveable regions, this representation relatively reduces the number of collisions with objects by up to 17% as compared to freespace-centric motion planners.

• Self-improving Multiplane-to-layer Images for Novel View Synthesis, WACV2023 | [code]

  We present a new method for lightweight novel-view synthesis that generalizes to an arbitrary forward-facing scene. Recent approaches are computationally expensive, require per-scene optimization, or produce a memory-expensive representation. We start by representing the scene with a set of fronto-parallel semitransparent planes and afterward convert them to deformable layers in an end-to-end manner. Additionally, we employ a feed-forward refinement procedure that corrects the estimated representation by aggregating information from input views. Our method does not require fine-tuning when a new scene is processed and can handle an arbitrary number of views without restrictions. Experimental results show that our approach surpasses recent models in terms of common metrics and human evaluation, with the noticeable advantage in inference speed and compactness of the inferred layered geometry, see this https URL

• Uncertainty-Driven Active Vision for Implicit Scene Reconstruction | [code]

  Multi-view implicit scene reconstruction methods have become increasingly popular due to their ability to represent complex scene details. Recent efforts have been devoted to improving the representation of input information and to reducing the number of views required to obtain high quality reconstructions. Yet, perhaps surprisingly, the study of which views to select to maximally improve scene understanding remains largely unexplored. We propose an uncertainty-driven active vision approach for implicit scene reconstruction, which leverages occupancy uncertainty accumulated across the scene using volume rendering to select the next view to acquire. To this end, we develop an occupancy-based reconstruction method which accurately represents scenes using either 2D or 3D supervision. We evaluate our proposed approach on the ABC dataset and the in the wild CO3D dataset, and show that: (1) we are able to obtain high quality state-of-the-art occupancy reconstructions; (2) our perspective conditioned uncertainty definition is effective to drive improvements in next best view selection and outperforms strong baseline approaches; and (3) we can further improve shape understanding by performing a gradient-based search on the view selection candidates. Overall, our results highlight the importance of view selection for implicit scene reconstruction, making it a promising avenue to explore further.

• NARF22: Neural Articulated Radiance Fields for Configuration-Aware Rendering, IROS2022 | [code]

  Articulated objects pose a unique challenge for robotic perception and manipulation. Their increased number of degrees-of-freedom makes tasks such as localization computationally difficult, while also making the process of real-world dataset collection unscalable. With the aim of addressing these scalability issues, we propose Neural Articulated Radiance Fields (NARF22), a pipeline which uses a fully-differentiable, configuration-parameterized Neural Radiance Field (NeRF) as a means of providing high quality renderings of articulated objects. NARF22 requires no explicit knowledge of the object structure at inference time. We propose a two-stage parts-based training mechanism which allows the object rendering models to generalize well across the configuration space even if the underlying training data has as few as one configuration represented. We demonstrate the efficacy of NARF22 by training configurable renderers on a real-world articulated tool dataset collected via a Fetch mobile manipulation robot. We show the applicability of the model to gradient-based inference methods through a configuration estimation and 6 degree-of-freedom pose refinement task. The project webpage is available at: this https URL.

• Probabilistic Volumetric Fusion for Dense Monocular SLAM | [code]

  We present a novel method to reconstruct 3D scenes from images by leveraging deep dense monocular SLAM and fast uncertainty propagation. The proposed approach is able to 3D reconstruct scenes densely, accurately, and in real-time while being robust to extremely noisy depth estimates coming from dense monocular SLAM. Differently from previous approaches, that either use ad-hoc depth filters, or that estimate the depth uncertainty from RGB-D cameras' sensor models, our probabilistic depth uncertainty derives directly from the information matrix of the underlying bundle adjustment problem in SLAM. We show that the resulting depth uncertainty provides an excellent signal to weight the depth-maps for volumetric fusion. Without our depth uncertainty, the resulting mesh is noisy and with artifacts, while our approach generates an accurate 3D mesh with significantly fewer artifacts. We provide results on the challenging Euroc dataset, and show that our approach achieves 92% better accuracy than directly fusing depths from monocular SLAM, and up to 90% improvements compared to the best competing approach.

• SinGRAV: Learning a Generative Radiance Volume from a Single Natural Scene | [code]

  We present a 3D generative model for general natural scenes. Lacking necessary volumes of 3D data characterizing the target scene, we propose to learn from a single scene. Our key insight is that a natural scene often contains multiple constituents whose geometry, texture, and spatial arrangements follow some clear patterns, but still exhibit rich variations over different regions within the same scene. This suggests localizing the learning of a generative model on substantial local regions. Hence, we exploit a multi-scale convolutional network, which possesses the spatial locality bias in nature, to learn from the statistics of local regions at multiple scales within a single scene. In contrast to existing methods, our learning setup bypasses the need to collect data from many homogeneous 3D scenes for learning common features. We coin our method SinGRAV, for learning a Generative RAdiance Volume from a Single natural scene. We demonstrate the ability of SinGRAV in generating plausible and diverse variations from a single scene, the merits of SinGRAV over state-of-the-art generative neural scene methods, as well as the versatility of SinGRAV by its use in a variety of applications, spanning 3D scene editing, composition, and animation. Code and data will be released to facilitate further research.

• IntrinsicNeRF: Learning Intrinsic Neural Radiance Fields for Editable Novel View Synthesis | [code]

  We present intrinsic neural radiance fields, dubbed IntrinsicNeRF, that introduce intrinsic decomposition into the NeRF-based~\cite{mildenhall2020nerf} neural rendering method and can perform editable novel view synthesis in room-scale scenes while existing inverse rendering combined with neural rendering methods~\cite{zhang2021physg, zhang2022modeling} can only work on object-specific scenes. Given that intrinsic decomposition is a fundamentally ambiguous and under-constrained inverse problem, we propose a novel distance-aware point sampling and adaptive reflectance iterative clustering optimization method that enables IntrinsicNeRF with traditional intrinsic decomposition constraints to be trained in an unsupervised manner, resulting in temporally consistent intrinsic decomposition results. To cope with the problem of different adjacent instances of similar reflectance in a scene being incorrectly clustered together, we further propose a hierarchical clustering method with coarse-to-fine optimization to obtain a fast hierarchical indexing representation. It enables compelling real-time augmented reality applications such as scene recoloring, material editing, and illumination variation. Extensive experiments on Blender Object and Replica Scene demonstrate that we can obtain high-quality, consistent intrinsic decomposition results and high-fidelity novel view synthesis even for challenging sequences. Code and data are available on the project webpage: this https URL.

• Unsupervised Multi-View Object Segmentation Using Radiance Field Propagation, NeurIPS2022 | [code]

  We present radiance field propagation (RFP), a novel approach to segmenting objects in 3D during reconstruction given only unlabeled multi-view images of a scene. RFP is derived from emerging neural radiance field-based techniques, which jointly encodes semantics with appearance and geometry. The core of our method is a novel propagation strategy for individual objects' radiance fields with a bidirectional photometric loss, enabling an unsupervised partitioning of a scene into salient or meaningful regions corresponding to different object instances. To better handle complex scenes with multiple objects and occlusions, we further propose an iterative expectation-maximization algorithm to refine object masks. To the best of our knowledge, RFP is the first unsupervised approach for tackling 3D scene object segmentation for neural radiance field (NeRF) without any supervision, annotations, or other cues such as 3D bounding boxes and prior knowledge of object class. Experiments demonstrate that RFP achieves feasible segmentation results that are more accurate than previous unsupervised image/scene segmentation approaches, and are comparable to existing supervised NeRF-based methods. The segmented object representations enable individual 3D object editing operations.

• MonoNHR: Monocular Neural Human Renderer | [code]

  Existing neural human rendering methods struggle with a single image input due to the lack of information in invisible areas and the depth ambiguity of pixels in visible areas. In this regard, we propose Monocular Neural Human Renderer (MonoNHR), a novel approach that renders robust free-viewpoint images of an arbitrary human given only a single image. MonoNHR is the first method that (i) renders human subjects never seen during training in a monocular setup, and (ii) is trained in a weakly-supervised manner without geometry supervision. First, we propose to disentangle 3D geometry and texture features and to condition the texture inference on the 3D geometry features. Second, we introduce a Mesh Inpainter module that inpaints the occluded parts exploiting human structural priors such as symmetry. Experiments on ZJU-MoCap, AIST, and HUMBI datasets show that our approach significantly outperforms the recent methods adapted to the monocular case.

• NeRF: Neural Radiance Field in 3D Vision, A Comprehensive Review | [code]

  Neural Radiance Field (NeRF), a new novel view synthesis with implicit scene representation has taken the field of Computer Vision by storm. As a novel view synthesis and 3D reconstruction method, NeRF models find applications in robotics, urban mapping, autonomous navigation, virtual reality/augmented reality, and more. Since the original paper by Mildenhall et al., more than 250 preprints were published, with more than 100 eventually being accepted in tier one Computer Vision Conferences. Given NeRF popularity and the current interest in this research area, we believe it necessary to compile a comprehensive survey of NeRF papers from the past two years, which we organized into both architecture, and application based taxonomies. We also provide an introduction to the theory of NeRF based novel view synthesis, and a benchmark comparison of the performance and speed of key NeRF models. By creating this survey, we hope to introduce new researchers to NeRF, provide a helpful reference for influential works in this field, as well as motivate future research directions with our discussion section.

Sep25 - Oct1, 2022

• Structure-Aware NeRF without Posed Camera via Epipolar Constraint | [code]

  The neural radiance field (NeRF) for realistic novel view synthesis requires camera poses to be pre-acquired by a structure-from-motion (SfM) approach. This two-stage strategy is not convenient to use and degrades the performance because the error in the pose extraction can propagate to the view synthesis. We integrate the pose extraction and view synthesis into a single end-to-end procedure so they can benefit from each other. For training NeRF models, only RGB images are given, without pre-known camera poses. The camera poses are obtained by the epipolar constraint in which the identical feature in different views has the same world coordinates transformed from the local camera coordinates according to the extracted poses. The epipolar constraint is jointly optimized with pixel color constraint. The poses are represented by a CNN-based deep network, whose input is the related frames. This joint optimization enables NeRF to be aware of the scene's structure that has an improved generalization performance. Extensive experiments on a variety of scenes demonstrate the effectiveness of the proposed approach. Code is available at this https URL.

• SCI: A spectrum concentrated implicit neural compression for biomedical data | [code]

  Massive collection and explosive growth of the huge amount of medical data, demands effective compression for efficient storage, transmission and sharing. Readily available visual data compression techniques have been studied extensively but tailored for nature images/videos, and thus show limited performance on medical data which are of different characteristics. Emerging implicit neural representation (INR) is gaining momentum and demonstrates high promise for fitting diverse visual data in target-data-specific manner, but a general compression scheme covering diverse medical data is so far absent. To address this issue, we firstly derive a mathematical explanation for INR's spectrum concentration property and an analytical insight on the design of compression-oriented INR architecture. Further, we design a funnel shaped neural network capable of covering broad spectrum of complex medical data and achieving high compression ratio. Based on this design, we conduct compression via optimization under given budget and propose an adaptive compression approach SCI, which adaptively partitions the target data into blocks matching the concentrated spectrum envelop of the adopted INR, and allocates parameter with high representation accuracy under given compression ratio. The experiments show SCI's superior performance over conventional techniques and wide applicability across diverse medical data.

• Improving 3D-aware Image Synthesis with A Geometry-aware Discriminator, NeurIPS2022 | [code]

  3D-aware image synthesis aims at learning a generative model that can render photo-realistic 2D images while capturing decent underlying 3D shapes. A popular solution is to adopt the generative adversarial network (GAN) and replace the generator with a 3D renderer, where volume rendering with neural radiance field (NeRF) is commonly used. Despite the advancement of synthesis quality, existing methods fail to obtain moderate 3D shapes. We argue that, considering the two-player game in the formulation of GANs, only making the generator 3D-aware is not enough. In other words, displacing the generative mechanism only offers the capability, but not the guarantee, of producing 3D-aware images, because the supervision of the generator primarily comes from the discriminator. To address this issue, we propose GeoD through learning a geometry-aware discriminator to improve 3D-aware GANs. Concretely, besides differentiating real and fake samples from the 2D image space, the discriminator is additionally asked to derive the geometry information from the inputs, which is then applied as the guidance of the generator. Such a simple yet effective design facilitates learning substantially more accurate 3D shapes. Extensive experiments on various generator architectures and training datasets verify the superiority of GeoD over state-of-the-art alternatives. Moreover, our approach is registered as a general framework such that a more capable discriminator (i.e., with a third task of novel view synthesis beyond domain classification and geometry extraction) can further assist the generator with a better multi-view consistency.

• Understanding Pure CLIP Guidance for Voxel Grid NeRF Models | [code]

  We explore the task of text to 3D object generation using CLIP. Specifically, we use CLIP for guidance without access to any datasets, a setting we refer to as pure CLIP guidance. While prior work has adopted this setting, there is no systematic study of mechanics for preventing adversarial generations within CLIP. We illustrate how different image-based augmentations prevent the adversarial generation problem, and how the generated results are impacted. We test different CLIP model architectures and show that ensembling different models for guidance can prevent adversarial generations within bigger models and generate sharper results. Furthermore, we implement an implicit voxel grid model to show how neural networks provide an additional layer of regularization, resulting in better geometrical structure and coherency of generated objects. Compared to prior work, we achieve more coherent results with higher memory efficiency and faster training speeds.

• Distilling Style from Image Pairs for Global Forward and Inverse Tone Mapping, CVMP2022 | [code]

  Many image enhancement or editing operations, such as forward and inverse tone mapping or color grading, do not have a unique solution, but instead a range of solutions, each representing a different style. Despite this, existing learning-based methods attempt to learn a unique mapping, disregarding this style. In this work, we show that information about the style can be distilled from collections of image pairs and encoded into a 2- or 3-dimensional vector. This gives us not only an efficient representation but also an interpretable latent space for editing the image style. We represent the global color mapping between a pair of images as a custom normalizing flow, conditioned on a polynomial basis of the pixel color. We show that such a network is more effective than PCA or VAE at encoding image style in low-dimensional space and lets us obtain an accuracy close to 40 dB, which is about 7-10 dB improvement over the state-of-the-art methods.

• Sphere-Guided Training of Neural Implicit Surfaces | [code]

  In recent years, surface modeling via neural implicit functions has become one of the main techniques for multi-view 3D reconstruction. However, the state-of-the-art methods rely on the implicit functions to model an entire volume of the scene, leading to reduced reconstruction fidelity in the areas with thin objects or high-frequency details. To address that, we present a method for jointly training neural implicit surfaces alongside an auxiliary explicit shape representation, which acts as surface guide. In our approach, this representation encapsulates the surface region of the scene and enables us to boost the efficiency of the implicit function training by only modeling the volume in that region. We propose using a set of learnable spherical primitives as a learnable surface guidance since they can be efficiently trained alongside the neural surface function using its gradients. Our training pipeline consists of iterative updates of the spheres' centers using the gradients of the implicit function and then fine-tuning the latter to the updated surface region of the scene. We show that such modification to the training procedure can be plugged into several popular implicit reconstruction methods, improving the quality of the results over multiple 3D reconstruction benchmarks.

• Towards Multi-spatiotemporal-scale Generalized PDE Modeling | [code]

  Partial differential equations (PDEs) are central to describing complex physical system simulations. Their expensive solution techniques have led to an increased interest in deep neural network based surrogates. However, the practical utility of training such surrogates is contingent on their ability to model complex multi-scale spatio-temporal phenomena. Various neural network architectures have been proposed to target such phenomena, most notably Fourier Neural Operators (FNOs) which give a natural handle over local & global spatial information via parameterization of different Fourier modes, and U-Nets which treat local and global information via downsampling and upsampling paths. However, generalizing across different equation parameters or different time-scales still remains a challenge. In this work, we make a comprehensive comparison between various FNO and U-Net like approaches on fluid mechanics problems in both vorticity-stream and velocity function form. For U-Nets, we transfer recent architectural improvements from computer vision, most notably from object segmentation and generative modeling. We further analyze the design considerations for using FNO layers to improve performance of U-Net architectures without major degradation of computational performance. Finally, we show promising results on generalization to different PDE parameters and time-scales with a single surrogate model.

• MonoNeuralFusion: Online Monocular Neural 3D Reconstruction with Geometric Priors | [code]

  High-fidelity 3D scene reconstruction from monocular videos continues to be challenging, especially for complete and fine-grained geometry reconstruction. The previous 3D reconstruction approaches with neural implicit representations have shown a promising ability for complete scene reconstruction, while their results are often over-smooth and lack enough geometric details. This paper introduces a novel neural implicit scene representation with volume rendering for high-fidelity online 3D scene reconstruction from monocular videos. For fine-grained reconstruction, our key insight is to incorporate geometric priors into both the neural implicit scene representation and neural volume rendering, thus leading to an effective geometry learning mechanism based on volume rendering optimization. Benefiting from this, we present MonoNeuralFusion to perform the online neural 3D reconstruction from monocular videos, by which the 3D scene geometry is efficiently generated and optimized during the on-the-fly 3D monocular scanning. The extensive comparisons with state-of-the-art approaches show that our MonoNeuralFusion consistently generates much better complete and fine-grained reconstruction results, both quantitatively and qualitatively.

• Implicit Neural Spatial Representations for Time-dependent PDEs | [code]

  Numerically solving partial differential equations (PDEs) often entails spatial and temporal discretizations. Traditional methods (e.g., finite difference, finite element, smoothed-particle hydrodynamics) frequently adopt explicit spatial discretizations, such as grids, meshes, and point clouds, where each degree-of-freedom corresponds to a location in space. While these explicit spatial correspondences are intuitive to model and understand, these representations are not necessarily optimal for accuracy, memory-usage, or adaptivity. In this work, we explore implicit neural representation as an alternative spatial discretization, where spatial information is implicitly stored in the neural network weights. With implicit neural spatial representation, PDE-constrained time-stepping translates into updating neural network weights, which naturally integrates with commonly adopted optimization time integrators. We validate our approach on a variety of classic PDEs with examples involving large elastic deformations, turbulent fluids, and multiscale phenomena. While slower to compute than traditional representations, our approach exhibits higher accuracy, lower memory consumption, and dynamically adaptive allocation of degrees of freedom without complex remeshing.

• Continuous PDE Dynamics Forecasting with Implicit Neural Representations | [code]

  Effective data-driven PDE forecasting methods often rely on fixed spatial and / or temporal discretizations. This raises limitations in real-world applications like weather prediction where flexible extrapolation at arbitrary spatiotemporal locations is required. We address this problem by introducing a new data-driven approach, DINo, that models a PDE's flow with continuous-time dynamics of spatially continuous functions. This is achieved by embedding spatial observations independently of their discretization via Implicit Neural Representations in a small latent space temporally driven by a learned ODE. This separate and flexible treatment of time and space makes DINo the first data-driven model to combine the following advantages. It extrapolates at arbitrary spatial and temporal locations; it can learn from sparse irregular grids or manifolds; at test time, it generalizes to new grids or resolutions. DINo outperforms alternative neural PDE forecasters in a variety of challenging generalization scenarios on representative PDE systems.

• SymmNeRF: Learning to Explore Symmetry Prior for Single-View View Synthesis, ACCV2022 | [code]

  We study the problem of novel view synthesis of objects from a single image. Existing methods have demonstrated the potential in single-view view synthesis. However, they still fail to recover the fine appearance details, especially in self-occluded areas. This is because a single view only provides limited information. We observe that manmade objects usually exhibit symmetric appearances, which introduce additional prior knowledge. Motivated by this, we investigate the potential performance gains of explicitly embedding symmetry into the scene representation. In this paper, we propose SymmNeRF, a neural radiance field (NeRF) based framework that combines local and global conditioning under the introduction of symmetry priors. In particular, SymmNeRF takes the pixel-aligned image features and the corresponding symmetric features as extra inputs to the NeRF, whose parameters are generated by a hypernetwork. As the parameters are conditioned on the image-encoded latent codes, SymmNeRF is thus scene-independent and can generalize to new scenes. Experiments on synthetic and realworld datasets show that SymmNeRF synthesizes novel views with more details regardless of the pose transformation, and demonstrates good generalization when applied to unseen objects. Code is available at: this https URL.

• Towards General-Purpose Representation Learning of Polygonal Geometries, GeoInformatica | [code]

  Neural network representation learning for spatial data is a common need for geographic artificial intelligence (GeoAI) problems. In recent years, many advancements have been made in representation learning for points, polylines, and networks, whereas little progress has been made for polygons, especially complex polygonal geometries. In this work, we focus on developing a general-purpose polygon encoding model, which can encode a polygonal geometry (with or without holes, single or multipolygons) into an embedding space. The result embeddings can be leveraged directly (or finetuned) for downstream tasks such as shape classification, spatial relation prediction, and so on. To achieve model generalizability guarantees, we identify a few desirable properties: loop origin invariance, trivial vertex invariance, part permutation invariance, and topology awareness. We explore two different designs for the encoder: one derives all representations in the spatial domain; the other leverages spectral domain representations. For the spatial domain approach, we propose ResNet1D, a 1D CNN-based polygon encoder, which uses circular padding to achieve loop origin invariance on simple polygons. For the spectral domain approach, we develop NUFTspec based on Non-Uniform Fourier Transformation (NUFT), which naturally satisfies all the desired properties. We conduct experiments on two tasks: 1) shape classification based on MNIST; 2) spatial relation prediction based on two new datasets - DBSR-46K and DBSR-cplx46K. Our results show that NUFTspec and ResNet1D outperform multiple existing baselines with significant margins. While ResNet1D suffers from model performance degradation after shape-invariance geometry modifications, NUFTspec is very robust to these modifications due to the nature of the NUFT.

• City-scale Incremental Neural Mapping with Three-layer Sampling and Panoptic Representation | [code]

  Neural implicit representations are drawing a lot of attention from the robotics community recently, as they are expressive, continuous and compact. However, city-scale incremental implicit dense mapping based on sparse LiDAR input is still an under-explored challenge. To this end,we successfully build the first city-scale incremental neural mapping system with a panoptic representation that consists of both environment-level and instance-level modelling. Given a stream of sparse LiDAR point cloud, it maintains a dynamic generative model that maps 3D coordinates to signed distance field (SDF) values. To address the difficulty of representing geometric information at different levels in city-scale space, we propose a tailored three-layer sampling strategy to dynamically sample the global, local and near-surface domains. Meanwhile, to realize high fidelity mapping, category-specific prior is introduced to better model the geometric details, leading to a panoptic representation. We evaluate on the public SemanticKITTI dataset and demonstrate the significance of the newly proposed three-layer sampling strategy and panoptic representation, using both quantitative and qualitative results. Codes and data will be publicly available.

• 360FusionNeRF: Panoramic Neural Radiance Fields with Joint Guidance | [code]

  We present a method to synthesize novel views from a single 360∘ panorama image based on the neural radiance field (NeRF). Prior studies in a similar setting rely on the neighborhood interpolation capability of multi-layer perceptions to complete missing regions caused by occlusion, which leads to artifacts in their predictions. We propose 360FusionNeRF, a semi-supervised learning framework where we introduce geometric supervision and semantic consistency to guide the progressive training process. Firstly, the input image is re-projected to 360∘ images, and auxiliary depth maps are extracted at other camera positions. The depth supervision, in addition to the NeRF color guidance, improves the geometry of the synthesized views. Additionally, we introduce a semantic consistency loss that encourages realistic renderings of novel views. We extract these semantic features using a pre-trained visual encoder such as CLIP, a Vision Transformer trained on hundreds of millions of diverse 2D photographs mined from the web with natural language supervision. Experiments indicate that our proposed method can produce plausible completions of unobserved regions while preserving the features of the scene. When trained across various scenes, 360FusionNeRF consistently achieves the state-of-the-art performance when transferring to synthetic Structured3D dataset (PSNR5%, SSIM3% LPIPS13%), real-world Matterport3D dataset (PSNR3%, SSIM3% LPIPS9%) and Replica360 dataset (PSNR8%, SSIM2% LPIPS~18%).

• Orbeez-SLAM: A Real-time Monocular Visual SLAM with ORB Features and NeRF-realized Mapping | [code]

  A spatial AI that can perform complex tasks through visual signals and cooperate with humans is highly anticipated. To achieve this, we need a visual SLAM that easily adapts to new scenes without pre-training and generates dense maps for downstream tasks in real-time. None of the previous learning-based and non-learning-based visual SLAMs satisfy all needs due to the intrinsic limitations of their components. In this work, we develop a visual SLAM named Orbeez-SLAM, which successfully collaborates with implicit neural representation (NeRF) and visual odometry to achieve our goals. Moreover, Orbeez-SLAM can work with the monocular camera since it only needs RGB inputs, making it widely applicable to the real world. We validate its effectiveness on various challenging benchmarks. Results show that our SLAM is up to 800x faster than the strong baseline with superior rendering outcomes.

• Enforcing safety for vision-based controllers via Control Barrier Functions and Neural Radiance Fields | [code]

  To navigate complex environments, robots must increasingly use high-dimensional visual feedback (e.g. images) for control. However, relying on high-dimensional image data to make control decisions raises important questions; particularly, how might we prove the safety of a visual-feedback controller? Control barrier functions (CBFs) are powerful tools for certifying the safety of feedback controllers in the state-feedback setting, but CBFs have traditionally been poorly-suited to visual feedback control due to the need to predict future observations in order to evaluate the barrier function. In this work, we solve this issue by leveraging recent advances in neural radiance fields (NeRFs), which learn implicit representations of 3D scenes and can render images from previously-unseen camera perspectives, to provide single-step visual foresight for a CBF-based controller. This novel combination is able to filter out unsafe actions and intervene to preserve safety. We demonstrate the effect of our controller in real-time simulation experiments where it successfully prevents the robot from taking dangerous actions.

• Efficient View Path Planning for Autonomous Implicit Reconstruction | [code]

  Implicit neural representations have shown promising potential for the 3D scene reconstruction. Recent work applies it to autonomous 3D reconstruction by learning information gain for view path planning. Effective as it is, the computation of the information gain is expensive, and compared with that using volumetric representations, collision checking using the implicit representation for a 3D point is much slower. In the paper, we propose to 1) leverage a neural network as an implicit function approximator for the information gain field and 2) combine the implicit fine-grained representation with coarse volumetric representations to improve efficiency. Further with the improved efficiency, we propose a novel informative path planning based on a graph-based planner. Our method demonstrates significant improvements in the reconstruction quality and planning efficiency compared with autonomous reconstructions with implicit and explicit representations. We deploy the method on a real UAV and the results show that our method can plan informative views and reconstruct a scene with high quality.

• WaterNeRF: Neural Radiance Fields for Underwater Scenes | [code]

  Underwater imaging is a critical task performed by marine robots for a wide range of applications including aquaculture, marine infrastructure inspection, and environmental monitoring. However, water column effects, such as attenuation and backscattering, drastically change the color and quality of imagery captured underwater. Due to varying water conditions and range-dependency of these effects, restoring underwater imagery is a challenging problem. This impacts downstream perception tasks including depth estimation and 3D reconstruction. In this paper, we advance state-of-the-art in neural radiance fields (NeRFs) to enable physics-informed dense depth estimation and color correction. Our proposed method, WaterNeRF, estimates parameters of a physics-based model for underwater image formation, leading to a hybrid data-driven and model-based solution. After determining the scene structure and radiance field, we can produce novel views of degraded as well as corrected underwater images, along with dense depth of the scene. We evaluate the proposed method qualitatively and quantitatively on a real underwater dataset.

• Neural Global Illumination: Interactive Indirect Illumination Prediction under Dynamic Area Lights, TVCG2022 | [code]

  We propose neural global illumination, a novel method for fast rendering full global illumination in static scenes with dynamic viewpoint and area lighting. The key idea of our method is to utilize a deep rendering network to model the complex mapping from each shading point to global illumination. To efficiently learn the mapping, we propose a neural-network-friendly input representation including attributes of each shading point, viewpoint information, and a combinational lighting representation that enables high-quality fitting with a compact neural network. To synthesize high-frequency global illumination effects, we transform the low-dimension input to higher-dimension space by positional encoding and model the rendering network as a deep fully-connected network. Besides, we feed a screen-space neural buffer to our rendering network to share global information between objects in the screen-space to each shading point. We have demonstrated our neural global illumination method in rendering a wide variety of scenes exhibiting complex and all-frequency global illumination effects such as multiple-bounce glossy interreflection, color bleeding, and caustics.

• Baking in the Feature: Accelerating Volumetric Segmentation by Rendering Feature Maps | [code]

  Methods have recently been proposed that densely segment 3D volumes into classes using only color images and expert supervision in the form of sparse semantically annotated pixels. While impressive, these methods still require a relatively large amount of supervision and segmenting an object can take several minutes in practice. Such systems typically only optimize their representation on the particular scene they are fitting, without leveraging any prior information from previously seen images. In this paper, we propose to use features extracted with models trained on large existing datasets to improve segmentation performance. We bake this feature representation into a Neural Radiance Field (NeRF) by volumetrically rendering feature maps and supervising on features extracted from each input image. We show that by baking this representation into the NeRF, we make the subsequent classification task much easier. Our experiments show that our method achieves higher segmentation accuracy with fewer semantic annotations than existing methods over a wide range of scenes.

Sep18 - Sep24, 2022

• Local_INN: Implicit Map Representation and Localization with Invertible Neural Networks | [code]

  Robot localization is an inverse problem of finding a robot's pose using a map and sensor measurements. In recent years, Invertible Neural Networks (INNs) have successfully solved ambiguous inverse problems in various fields. This paper proposes a framework that solves the localization problem with INN. We design an INN that provides implicit map representation in the forward path and localization in the inverse path. By sampling the latent space in evaluation, Local_INN outputs robot poses with covariance, which can be used to estimate the uncertainty. We show that the localization performance of Local_INN is on par with current methods with much lower latency. We show detailed 2D and 3D map reconstruction from Local_INN using poses exterior to the training set. We also provide a global localization algorithm using Local_INN to tackle the kidnapping problem.

• NeRF-Loc: Transformer-Based Object Localization Within Neural Radiance Fields | [code]

  Neural Radiance Fields (NeRFs) have been successfully used for scene representation. Recent works have also developed robotic navigation and manipulation systems using NeRF-based environment representations. As object localization is the foundation for many robotic applications, to further unleash the potential of NeRFs in robotic systems, we study object localization within a NeRF scene. We propose a transformer-based framework NeRF-Loc to extract 3D bounding boxes of objects in NeRF scenes. NeRF-Loc takes a pre-trained NeRF model and camera view as input, and produces labeled 3D bounding boxes of objects as output. Concretely, we design a pair of paralleled transformer encoder branches, namely the coarse stream and the fine stream, to encode both the context and details of target objects. The encoded features are then fused together with attention layers to alleviate ambiguities for accurate object localization. We have compared our method with the conventional transformer-based method and our method achieves better performance. In addition, we also present the first NeRF samples-based object localization benchmark NeRFLocBench.

• SG-SRNs: Superpixel-Guided Scene Representation Networks, SignalProcessingLetters | [code]

  Recently, Scene Representation Networks (SRNs) have attracted increasing attention in computer vision, due to their continuous and light-weight scene representation ability. However, SRNs generally perform poorly on low-texture image regions. Addressing this problem, we propose superpixel-guided scene representation networks in this paper, called SG-SRNs, consisting of a backbone module (SRNs), a superpixel segmentation module, and a superpixel regularization module. In the proposed method, except for the novel view synthesis task, the task of representation-aware superpixel segmentation mask generation is realized by the proposed superpixel segmentation module. Then, the superpixel regularization module utilizes the superpixel segmentation mask to guide the backbone to be learned in a locally smooth way, and optimizes the scene representations of the local regions to indirectly alleviate the structure distortion of low-texture regions in a self-supervised manner. Extensive experimental results on both our constructed datasets and the public Synthetic-NeRF dataset demonstrated that the proposed SG-SRNs achieved a significantly better 3D structure representing performance.

• PNeRF: Probabilistic Neural Scene Representations for Uncertain 3D Visual Mapping, ICRA2023 | [code]

  Recently neural scene representations have provided very impressive results for representing 3D scenes visually, however, their study and progress have mainly been limited to visualization of virtual models in computer graphics or scene reconstruction in computer vision without explicitly accounting for sensor and pose uncertainty. Using this novel scene representation in robotics applications, however, would require accounting for this uncertainty in the neural map. The aim of this paper is therefore to propose a novel method for training {\em probabilistic neural scene representations} with uncertain training data that could enable the inclusion of these representations in robotics applications. Acquiring images using cameras or depth sensors contains inherent uncertainty, and furthermore, the camera poses used for learning a 3D model are also imperfect. If these measurements are used for training without accounting for their uncertainty, then the resulting models are non-optimal, and the resulting scene representations are likely to contain artifacts such as blur and un-even geometry. In this work, the problem of uncertainty integration to the learning process is investigated by focusing on training with uncertain information in a probabilistic manner. The proposed method involves explicitly augmenting the training likelihood with an uncertainty term such that the learnt probability distribution of the network is minimized with respect to the training uncertainty. It will be shown that this leads to more accurate image rendering quality, in addition to more precise and consistent geometry. Validation has been carried out on both synthetic and real datasets showing that the proposed approach outperforms state-of-the-art methods. The results show notably that the proposed method is capable of rendering novel high-quality views even when the training data is limited.

• How Does It Feel? Self-Supervised Costmap Learning for Off-Road Vehicle Traversability | [code]

  Estimating terrain traversability in off-road environments requires reasoning about complex interaction dynamics between the robot and these terrains. However, it is challenging to build an accurate physics model, or create informative labels to learn a model in a supervised manner, for these interactions. We propose a method that learns to predict traversability costmaps by combining exteroceptive environmental information with proprioceptive terrain interaction feedback in a self-supervised manner. Additionally, we propose a novel way of incorporating robot velocity in the costmap prediction pipeline. We validate our method in multiple short and large-scale navigation tasks on a large, autonomous all-terrain vehicle (ATV) on challenging off-road terrains, and demonstrate ease of integration on a separate large ground robot. Our short-scale navigation results show that using our learned costmaps leads to overall smoother navigation, and provides the robot with a more fine-grained understanding of the interactions between the robot and different terrain types, such as grass and gravel. Our large-scale navigation trials show that we can reduce the number of interventions by up to 57% compared to an occupancy-based navigation baseline in challenging off-road courses ranging from 400 m to 3150 m.

• Edge-oriented Implicit Neural Representation with Channel Tuning | [code]

  Implicit neural representation, which expresses an image as a continuous function rather than a discrete grid form, is widely used for image processing. Despite its outperforming results, there are still remaining limitations on restoring clear shapes of a given signal such as the edges of an image. In this paper, we propose Gradient Magnitude Adjustment algorithm which calculates the gradient of an image for training the implicit representation. In addition, we propose Edge-oriented Representation Network (EoREN) that can reconstruct the image with clear edges by fitting gradient information (Edge-oriented module). Furthermore, we add Channel-tuning module to adjust the distribution of given signals so that it solves a chronic problem of fitting gradients. By separating backpropagation paths of the two modules, EoREN can learn true color of the image without hindering the role for gradients. We qualitatively show that our model can reconstruct complex signals and demonstrate general reconstruction ability of our model with quantitative results.

• Fast Disparity Estimation from a Single Compressed Light Field Measurement | [code]

  The abundant spatial and angular information from light fields has allowed the development of multiple disparity estimation approaches. However, the acquisition of light fields requires high storage and processing cost, limiting the use of this technology in practical applications. To overcome these drawbacks, the compressive sensing (CS) theory has allowed the development of optical architectures to acquire a single coded light field measurement. This measurement is decoded using an optimization algorithm or deep neural network that requires high computational costs. The traditional approach for disparity estimation from compressed light fields requires first recovering the entire light field and then a post-processing step, thus requiring long times. In contrast, this work proposes a fast disparity estimation from a single compressed measurement by omitting the recovery step required in traditional approaches. Specifically, we propose to jointly optimize an optical architecture for acquiring a single coded light field snapshot and a convolutional neural network (CNN) for estimating the disparity maps. Experimentally, the proposed method estimates disparity maps comparable with those obtained from light fields reconstructed using deep learning approaches. Furthermore, the proposed method is 20 times faster in training and inference than the best method that estimates the disparity from reconstructed light fields.

• FNeVR: Neural Volume Rendering for Face Animation | [code]

  Face animation, one of the hottest topics in computer vision, has achieved a promising performance with the help of generative models. However, it remains a critical challenge to generate identity preserving and photo-realistic images due to the sophisticated motion deformation and complex facial detail modeling. To address these problems, we propose a Face Neural Volume Rendering (FNeVR) network to fully explore the potential of 2D motion warping and 3D volume rendering in a unified framework. In FNeVR, we design a 3D Face Volume Rendering (FVR) module to enhance the facial details for image rendering. Specifically, we first extract 3D information with a well-designed architecture, and then introduce an orthogonal adaptive ray-sampling module for efficient rendering. We also design a lightweight pose editor, enabling FNeVR to edit the facial pose in a simple yet effective way. Extensive experiments show that our FNeVR obtains the best overall quality and performance on widely used talking-head benchmarks.

• PREF: Predictability Regularized Neural Motion Fields, ECCV2022(oral) | [code]

  Knowing the 3D motions in a dynamic scene is essential to many vision applications. Recent progress is mainly focused on estimating the activity of some specific elements like humans. In this paper, we leverage a neural motion field for estimating the motion of all points in a multiview setting. Modeling the motion from a dynamic scene with multiview data is challenging due to the ambiguities in points of similar color and points with time-varying color. We propose to regularize the estimated motion to be predictable. If the motion from previous frames is known, then the motion in the near future should be predictable. Therefore, we introduce a predictability regularization by first conditioning the estimated motion on latent embeddings, then by adopting a predictor network to enforce predictability on the embeddings. The proposed framework PREF (Predictability REgularized Fields) achieves on par or better results than state-of-the-art neural motion field-based dynamic scene representation methods, while requiring no prior knowledge of the scene.

• wildNeRF: Complete view synthesis of in-the-wild dynamic scenes captured using sparse monocular data | [code]

  We present a novel neural radiance model that is trainable in a self-supervised manner for novel-view synthesis of dynamic unstructured scenes. Our end-to-end trainable algorithm learns highly complex, real-world static scenes within seconds and dynamic scenes with both rigid and non-rigid motion within minutes. By differentiating between static and motion-centric pixels, we create high-quality representations from a sparse set of images. We perform extensive qualitative and quantitative evaluation on existing benchmarks and set the state-of-the-art on performance measures on the challenging NVIDIA Dynamic Scenes Dataset. Additionally, we evaluate our model performance on challenging real-world datasets such as Cholec80 and SurgicalActions160.

• Loc-NeRF: Monte Carlo Localization using Neural Radiance Fields | [code]

  We present Loc-NeRF, a real-time vision-based robot localization approach that combines Monte Carlo localization and Neural Radiance Fields (NeRF). Our system uses a pre-trained NeRF model as the map of an environment and can localize itself in real-time using an RGB camera as the only exteroceptive sensor onboard the robot. While neural radiance fields have seen significant applications for visual rendering in computer vision and graphics, they have found limited use in robotics. Existing approaches for NeRF-based localization require both a good initial pose guess and significant computation, making them impractical for real-time robotics applications. By using Monte Carlo localization as a workhorse to estimate poses using a NeRF map model, Loc-NeRF is able to perform localization faster than the state of the art and without relying on an initial pose estimate. In addition to testing on synthetic data, we also run our system using real data collected by a Clearpath Jackal UGV and demonstrate for the first time the ability to perform real-time global localization with neural radiance fields. We make our code publicly available at this https URL.

• Density-aware NeRF Ensembles: Quantifying Predictive Uncertainty in Neural Radiance Fields | [code]

  We show that ensembling effectively quantifies model uncertainty in Neural Radiance Fields (NeRFs) if a density-aware epistemic uncertainty term is considered. The naive ensembles investigated in prior work simply average rendered RGB images to quantify the model uncertainty caused by conflicting explanations of the observed scene. In contrast, we additionally consider the termination probabilities along individual rays to identify epistemic model uncertainty due to a lack of knowledge about the parts of a scene unobserved during training. We achieve new state-of-the-art performance across established uncertainty quantification benchmarks for NeRFs, outperforming methods that require complex changes to the NeRF architecture and training regime. We furthermore demonstrate that NeRF uncertainty can be utilised for next-best view selection and model refinement.

• NeRF-SOS: Any-View Self-supervised Object Segmentation on Complex Scenes | [code]

  Neural volumetric representations have shown the potential that Multi-layer Perceptrons (MLPs) can be optimized with multi-view calibrated images to represent scene geometry and appearance, without explicit 3D supervision. Object segmentation can enrich many downstream applications based on the learned radiance field. However, introducing hand-crafted segmentation to define regions of interest in a complex real-world scene is non-trivial and expensive as it acquires per view annotation. This paper carries out the exploration of self-supervised learning for object segmentation using NeRF for complex real-world scenes. Our framework, called NeRF with Self-supervised Object Segmentation NeRF-SOS, couples object segmentation and neural radiance field to segment objects in any view within a scene. By proposing a novel collaborative contrastive loss in both appearance and geometry levels, NeRF-SOS encourages NeRF models to distill compact geometry-aware segmentation clusters from their density fields and the self-supervised pre-trained 2D visual features. The self-supervised object segmentation framework can be applied to various NeRF models that both lead to photo-realistic rendering results and convincing segmentation maps for both indoor and outdoor scenarios. Extensive results on the LLFF, Tank & Temple, and BlendedMVS datasets validate the effectiveness of NeRF-SOS. It consistently surpasses other 2D-based self-supervised baselines and predicts finer semantics masks than existing supervised counterparts. Please refer to the video on our project page for more details:this https URL.

• MeSLAM: Memory Efficient SLAM based on Neural Fields, SMC2022 | [code]

  Existing Simultaneous Localization and Mapping (SLAM) approaches are limited in their scalability due to growing map size in long-term robot operation. Moreover, processing such maps for localization and planning tasks leads to the increased computational resources required onboard. To address the problem of memory consumption in long-term operation, we develop a novel real-time SLAM algorithm, MeSLAM, that is based on neural field implicit map representation. It combines the proposed global mapping strategy, including neural networks distribution and region tracking, with an external odometry system. As a result, the algorithm is able to efficiently train multiple networks representing different map regions and track poses accurately in large-scale environments. Experimental results show that the accuracy of the proposed approach is comparable to the state-of-the-art methods (on average, 6.6 cm on TUM RGB-D sequences) and outperforms the baseline, iMAP∗. Moreover, the proposed SLAM approach provides the most compact-sized maps without details distortion (1.9 MB to store 57 m3) among the state-of-the-art SLAM approaches.

• Human Performance Modeling and Rendering via Neural Animated Mesh | [code]

  We have recently seen tremendous progress in the neural advances for photo-real human modeling and rendering. However, it's still challenging to integrate them into an existing mesh-based pipeline for downstream applications. In this paper, we present a comprehensive neural approach for high-quality reconstruction, compression, and rendering of human performances from dense multi-view videos. Our core intuition is to bridge the traditional animated mesh workflow with a new class of highly efficient neural techniques. We first introduce a neural surface reconstructor for high-quality surface generation in minutes. It marries the implicit volumetric rendering of the truncated signed distance field (TSDF) with multi-resolution hash encoding. We further propose a hybrid neural tracker to generate animated meshes, which combines explicit non-rigid tracking with implicit dynamic deformation in a self-supervised framework. The former provides the coarse warping back into the canonical space, while the latter implicit one further predicts the displacements using the 4D hash encoding as in our reconstructor. Then, we discuss the rendering schemes using the obtained animated meshes, ranging from dynamic texturing to lumigraph rendering under various bandwidth settings. To strike an intricate balance between quality and bandwidth, we propose a hierarchical solution by first rendering 6 virtual views covering the performer and then conducting occlusion-aware neural texture blending. We demonstrate the efficacy of our approach in a variety of mesh-based applications and photo-realistic free-view experiences on various platforms, i.e., inserting virtual human performances into real environments through mobile AR or immersively watching talent shows with VR headsets.

• LATITUDE: Robotic Global Localization with Truncated Dynamic Low-pass Filter in City-scale NeRF, ICRA2023 | [code]

  Neural Radiance Fields (NeRFs) have made great success in representing complex 3D scenes with high-resolution details and efficient memory. Nevertheless, current NeRF-based pose estimators have no initial pose prediction and are prone to local optima during optimization. In this paper, we present LATITUDE: Global Localization with Truncated Dynamic Low-pass Filter, which introduces a two-stage localization mechanism in city-scale NeRF. In place recognition stage, we train a regressor through images generated from trained NeRFs, which provides an initial value for global localization. In pose optimization stage, we minimize the residual between the observed image and rendered image by directly optimizing the pose on tangent plane. To avoid convergence to local optimum, we introduce a Truncated Dynamic Low-pass Filter (TDLF) for coarse-to-fine pose registration. We evaluate our method on both synthetic and real-world data and show its potential applications for high-precision navigation in large-scale city scenes. Codes and data will be publicly available at this https URL.

• Neural Implicit Surface Reconstruction using Imaging Sonar | [code]

  We present a technique for dense 3D reconstruction of objects using an imaging sonar, also known as forward-looking sonar (FLS). Compared to previous methods that model the scene geometry as point clouds or volumetric grids, we represent the geometry as a neural implicit function. Additionally, given such a representation, we use a differentiable volumetric renderer that models the propagation of acoustic waves to synthesize imaging sonar measurements. We perform experiments on real and synthetic datasets and show that our algorithm reconstructs high-fidelity surface geometry from multi-view FLS images at much higher quality than was possible with previous techniques and without suffering from their associated memory overhead.

• Uncertainty Guided Policy for Active Robotic 3D Reconstruction using Neural Radiance Fields, RAL2022 | [code]

  In this paper, we tackle the problem of active robotic 3D reconstruction of an object. In particular, we study how a mobile robot with an arm-held camera can select a favorable number of views to recover an object's 3D shape efficiently. Contrary to the existing solution to this problem, we leverage the popular neural radiance fields-based object representation, which has recently shown impressive results for various computer vision tasks. However, it is not straightforward to directly reason about an object's explicit 3D geometric details using such a representation, making the next-best-view selection problem for dense 3D reconstruction challenging. This paper introduces a ray-based volumetric uncertainty estimator, which computes the entropy of the weight distribution of the color samples along each ray of the object's implicit neural representation. We show that it is possible to infer the uncertainty of the underlying 3D geometry given a novel view with the proposed estimator. We then present a next-best-view selection policy guided by the ray-based volumetric uncertainty in neural radiance fields-based representations. Encouraging experimental results on synthetic and real-world data suggest that the approach presented in this paper can enable a new research direction of using an implicit 3D object representation for the next-best-view problem in robot vision applications, distinguishing our approach from the existing approaches that rely on explicit 3D geometric modeling.

• Implicit Neural Representations for Medical Imaging Segmentation, MICCAI2022 | [code]

  3D signals in medical imaging, such as CT scans, are usually parameterized as a discrete grid of voxels. For instance, existing state-of-the-art organ segmentation methods learn discrete segmentation maps. Unfortunately, the memory requirements of such methods grow cubically with increasing spatial resolution, which makes them unsuitable for processing high resolution scans. To overcome this, we design an Implicit Organ Segmentation Network (IOSNet) that utilizes continuous Implicit Neural Representations and has several useful properties. Firstly, the IOSNet decoder memory is roughly constant and independent of the spatial resolution since it parameterizes the segmentation map as a continuous function. Secondly, IOSNet converges much faster than discrete voxel based methods due to its ability to accurately segment organs irrespective of organ sizes, thereby alleviating size imbalance issues without requiring any auxiliary tricks. Thirdly, IOSNet naturally supports super-resolution (i.e. sampling at arbitrary resolutions during inference) due to its continuous learnt representations. Moreover, despite using a simple lightweight decoder, IOSNet consistently outperforms the discrete specialized segmentation architecture UNet. Hence, our approach demonstrates that Implicit Neural Representations are well-suited for medical imaging applications, especially for processing high-resolution 3D medical scans.

• ActiveNeRF: Learning where to See with Uncertainty Estimation | [code]

  Recently, Neural Radiance Fields (NeRF) has shown promising performances on reconstructing 3D scenes and synthesizing novel views from a sparse set of 2D images. Albeit effective, the performance of NeRF is highly influenced by the quality of training samples. With limited posed images from the scene, NeRF fails to generalize well to novel views and may collapse to trivial solutions in unobserved regions. This makes NeRF impractical under resource-constrained scenarios. In this paper, we present a novel learning framework, ActiveNeRF, aiming to model a 3D scene with a constrained input budget. Specifically, we first incorporate uncertainty estimation into a NeRF model, which ensures robustness under few observations and provides an interpretation of how NeRF understands the scene. On this basis, we propose to supplement the existing training set with newly captured samples based on an active learning scheme. By evaluating the reduction of uncertainty given new inputs, we select the samples that bring the most information gain. In this way, the quality of novel view synthesis can be improved with minimal additional resources. Extensive experiments validate the performance of our model on both realistic and synthetic scenes, especially with scarcer training data. Code will be released at \url{this https URL}.

Sep11 - Sep17, 2022

• iDF-SLAM: End-to-End RGB-D SLAM with Neural Implicit Mapping and Deep Feature Tracking | [code]

  We propose a novel end-to-end RGB-D SLAM, iDF-SLAM, which adopts a feature-based deep neural tracker as the front-end and a NeRF-style neural implicit mapper as the back-end. The neural implicit mapper is trained on-the-fly, while though the neural tracker is pretrained on the ScanNet dataset, it is also finetuned along with the training of the neural implicit mapper. Under such a design, our iDF-SLAM is capable of learning to use scene-specific features for camera tracking, thus enabling lifelong learning of the SLAM system. Both the training for the tracker and the mapper are self-supervised without introducing ground truth poses. We test the performance of our iDF-SLAM on the Replica and ScanNet datasets and compare the results to the two recent NeRF-based neural SLAM systems. The proposed iDF-SLAM demonstrates state-of-the-art results in terms of scene reconstruction and competitive performance in camera tracking.

• 3DMM-RF: Convolutional Radiance Fields for 3D Face Modeling | [code]

  Facial 3D Morphable Models are a main computer vision subject with countless applications and have been highly optimized in the last two decades. The tremendous improvements of deep generative networks have created various possibilities for improving such models and have attracted wide interest. Moreover, the recent advances in neural radiance fields, are revolutionising novel-view synthesis of known scenes. In this work, we present a facial 3D Morphable Model, which exploits both of the above, and can accurately model a subject's identity, pose and expression and render it in arbitrary illumination. This is achieved by utilizing a powerful deep style-based generator to overcome two main weaknesses of neural radiance fields, their rigidity and rendering speed. We introduce a style-based generative network that synthesizes in one pass all and only the required rendering samples of a neural radiance field. We create a vast labelled synthetic dataset of facial renders, and train the network on these data, so that it can accurately model and generalize on facial identity, pose and appearance. Finally, we show that this model can accurately be fit to "in-the-wild" facial images of arbitrary pose and illumination, extract the facial characteristics, and be used to re-render the face in controllable conditions.

• DevNet: Self-supervised Monocular Depth Learning via Density Volume Construction, ECCV2022 | [code]

  Self-supervised depth learning from monocular images normally relies on the 2D pixel-wise photometric relation between temporally adjacent image frames. However, they neither fully exploit the 3D point-wise geometric correspondences, nor effectively tackle the ambiguities in the photometric warping caused by occlusions or illumination inconsistency. To address these problems, this work proposes Density Volume Construction Network (DevNet), a novel self-supervised monocular depth learning framework, that can consider 3D spatial information, and exploit stronger geometric constraints among adjacent camera frustums. Instead of directly regressing the pixel value from a single image, our DevNet divides the camera frustum into multiple parallel planes and predicts the pointwise occlusion probability density on each plane. The final depth map is generated by integrating the density along corresponding rays. During the training process, novel regularization strategies and loss functions are introduced to mitigate photometric ambiguities and overfitting. Without obviously enlarging model parameters size or running time, DevNet outperforms several representative baselines on both the KITTI-2015 outdoor dataset and NYU-V2 indoor dataset. In particular, the root-mean-square-deviation is reduced by around 4% with DevNet on both KITTI-2015 and NYU-V2 in the task of depth estimation. Code is available at this https URL.

• Explicitly Controllable 3D-Aware Portrait Generation | [code]

  In contrast to the traditional avatar creation pipeline which is a costly process, contemporary generative approaches directly learn the data distribution from photographs. While plenty of works extend unconditional generative models and achieve some levels of controllability, it is still challenging to ensure multi-view consistency, especially in large poses. In this work, we propose a network that generates 3D-aware portraits while being controllable according to semantic parameters regarding pose, identity, expression and illumination. Our network uses neural scene representation to model 3D-aware portraits, whose generation is guided by a parametric face model that supports explicit control. While the latent disentanglement can be further enhanced by contrasting images with partially different attributes, there still exists noticeable inconsistency in non-face areas, e.g., hair and background, when animating expressions. Wesolve this by proposing a volume blending strategy in which we form a composite output by blending dynamic and static areas, with two parts segmented from the jointly learned semantic field. Our method outperforms prior arts in extensive experiments, producing realistic portraits with vivid expression in natural lighting when viewed from free viewpoints. It also demonstrates generalization ability to real images as well as out-of-domain data, showing great promise in real applications.

• StructNeRF: Neural Radiance Fields for Indoor Scenes with Structural Hints | [code]

  Neural Radiance Fields (NeRF) achieve photo-realistic view synthesis with densely captured input images. However, the geometry of NeRF is extremely under-constrained given sparse views, resulting in significant degradation of novel view synthesis quality. Inspired by self-supervised depth estimation methods, we propose StructNeRF, a solution to novel view synthesis for indoor scenes with sparse inputs. StructNeRF leverages the structural hints naturally embedded in multi-view inputs to handle the unconstrained geometry issue in NeRF. Specifically, it tackles the texture and non-texture regions respectively: a patch-based multi-view consistent photometric loss is proposed to constrain the geometry of textured regions; for non-textured ones, we explicitly restrict them to be 3D consistent planes. Through the dense self-supervised depth constraints, our method improves both the geometry and the view synthesis performance of NeRF without any additional training on external data. Extensive experiments on several real-world datasets demonstrate that StructNeRF surpasses state-of-the-art methods for indoor scenes with sparse inputs both quantitatively and qualitatively.

• Learning A Unified 3D Point Cloud for View Synthesis | [code]

  3D point cloud representation-based view synthesis methods have demonstrated effectiveness. However, existing methods usually synthesize novel views only from a single source view, and it is non-trivial to generalize them to handle multiple source views for pursuing higher reconstruction quality. In this paper, we propose a new deep learning-based view synthesis paradigm, which learns a unified 3D point cloud from different source views. Specifically, we first construct sub-point clouds by projecting source views to 3D space based on their depth maps. Then, we learn the unified 3D point cloud by adaptively fusing points at a local neighborhood defined on the union of the sub-point clouds. Besides, we also propose a 3D geometry-guided image restoration module to fill the holes and recover high-frequency details of the rendered novel views. Experimental results on three benchmark datasets demonstrate that our method outperforms state-of-the-art view synthesis methods to a large extent both quantitatively and visually.

• Self-Supervised Coordinate Projection Network for Sparse-View Computed Tomography | [code]

  In the present work, we propose a Self-supervised COordinate Projection nEtwork (SCOPE) to reconstruct the artifacts-free CT image from a single SV sinogram by solving the inverse tomography imaging problem. Compared with recent related works that solve similar problems using implicit neural representation network (INR), our essential contribution is an effective and simple re-projection strategy that pushes the tomography image reconstruction quality over supervised deep learning CT reconstruction works. The proposed strategy is inspired by the simple relationship between linear algebra and inverse problems. To solve the under-determined linear equation system, we first introduce INR to constrain the solution space via image continuity prior and achieve a rough solution. And secondly, we propose to generate a dense view sinogram that improves the rank of the linear equation system and produces a more stable CT image solution space. Our experiment results demonstrate that the re-projection strategy significantly improves the image reconstruction quality (+3 dB for PSNR at least). Besides, we integrate the recent hash encoding into our SCOPE model, which greatly accelerates the model training. Finally, we evaluate SCOPE in parallel and fan X-ray beam SVCT reconstruction tasks. Experimental results indicate that the proposed SCOPE model outperforms two latest INR-based methods and two well-popular supervised DL methods quantitatively and qualitatively.

• CU-Net: Efficient Point Cloud Color Upsampling Network | [code]

  Point cloud upsampling is necessary for Augmented Reality, Virtual Reality, and telepresence scenarios. Although the geometry upsampling is well studied to densify point cloud coordinates, the upsampling of colors has been largely overlooked. In this paper, we propose CU-Net, the first deep-learning point cloud color upsampling model. Leveraging a feature extractor based on sparse convolution and a color prediction module based on neural implicit function, CU-Net achieves linear time and space complexity. Therefore, CU-Net is theoretically guaranteed to be more efficient than most existing methods with quadratic complexity. Experimental results demonstrate that CU-Net can colorize a photo-realistic point cloud with nearly a million points in real time, while having better visual quality than baselines. Besides, CU-Net can adapt to an arbitrary upsampling ratio and unseen objects. Our source code will be released to the public soon.

Sep4 - Sep10, 2022

• PixTrack: Precise 6DoF Object Pose Tracking using NeRF Templates and Feature-metric Alignment | [code]

  We present PixTrack, a vision based object pose tracking framework using novel view synthesis and deep feature-metric alignment. Our evaluations demonstrate that our method produces highly accurate, robust, and jitter-free 6DoF pose estimates of objects in RGB images without the need of any data annotation or trajectory smoothing. Our method is also computationally efficient making it easy to have multi-object tracking with no alteration to our method and just using CPU multiprocessing.

• Implicit Full Waveform Inversion with Deep Neural Representation | [code]

  Full waveform inversion (FWI) commonly stands for the state-of-the-art approach for imaging subsurface structures and physical parameters, however, its implementation usually faces great challenges, such as building a good initial model to escape from local minima, and evaluating the uncertainty of inversion results. In this paper, we propose the implicit full waveform inversion (IFWI) algorithm using continuously and implicitly defined deep neural representations. Compared to FWI, which is sensitive to the initial model, IFWI benefits from the increased degrees of freedom with deep learning optimization, thus allowing to start from a random initialization, which greatly reduces the risk of non-uniqueness and being trapped in local minima. Both theoretical and experimental analyses indicates that, given a random initial model, IFWI is able to converge to the global minimum and produce a high-resolution image of subsurface with fine structures. In addition, uncertainty analysis of IFWI can be easily performed by approximating Bayesian inference with various deep learning approaches, which is analyzed in this paper by adding dropout neurons. Furthermore, IFWI has a certain degree of robustness and strong generalization ability that are exemplified in the experiments of various 2D geological models. With proper setup, IFWI can also be well suited for multi-scale joint geophysical inversion.

• 3D Textured Shape Recovery with Learned Geometric Priors | [code]

  3D textured shape recovery from partial scans is crucial for many real-world applications. Existing approaches have demonstrated the efficacy of implicit function representation, but they suffer from partial inputs with severe occlusions and varying object types, which greatly hinders their application value in the real world. This technical report presents our approach to address these limitations by incorporating learned geometric priors. To this end, we generate a SMPL model from learned pose prediction and fuse it into the partial input to add prior knowledge of human bodies. We also propose a novel completeness-aware bounding box adaptation for handling different levels of scales and partialness of partial scans.

• SIRA: Relightable Avatars from a Single Image | [code]

  Recovering the geometry of a human head from a single image, while factorizing the materials and illumination is a severely ill-posed problem that requires prior information to be solved. Methods based on 3D Morphable Models (3DMM), and their combination with differentiable renderers, have shown promising results. However, the expressiveness of 3DMMs is limited, and they typically yield over-smoothed and identity-agnostic 3D shapes limited to the face region. Highly accurate full head reconstructions have recently been obtained with neural fields that parameterize the geometry using multilayer perceptrons. The versatility of these representations has also proved effective for disentangling geometry, materials and lighting. However, these methods require several tens of input images. In this paper, we introduce SIRA, a method which, from a single image, reconstructs human head avatars with high fidelity geometry and factorized lights and surface materials. Our key ingredients are two data-driven statistical models based on neural fields that resolve the ambiguities of single-view 3D surface reconstruction and appearance factorization. Experiments show that SIRA obtains state of the art results in 3D head reconstruction while at the same time it successfully disentangles the global illumination, and the diffuse and specular albedos. Furthermore, our reconstructions are amenable to physically-based appearance editing and head model relighting.

• Neural Feature Fusion Fields: 3D Distillation of Self-Supervised 2D Image Representations, 3DV2022(oral) | [code]

  We present Neural Feature Fusion Fields (N3F), a method that improves dense 2D image feature extractors when the latter are applied to the analysis of multiple images reconstructible as a 3D scene. Given an image feature extractor, for example pre-trained using self-supervision, N3F uses it as a teacher to learn a student network defined in 3D space. The 3D student network is similar to a neural radiance field that distills said features and can be trained with the usual differentiable rendering machinery. As a consequence, N3F is readily applicable to most neural rendering formulations, including vanilla NeRF and its extensions to complex dynamic scenes. We show that our method not only enables semantic understanding in the context of scene-specific neural fields without the use of manual labels, but also consistently improves over the self-supervised 2D baselines. This is demonstrated by considering various tasks, such as 2D object retrieval, 3D segmentation, and scene editing, in diverse sequences, including long egocentric videos in the EPIC-KITCHENS benchmark.

• MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model | [code]

  Human motion modeling is important for many modern graphics applications, which typically require professional skills. In order to remove the skill barriers for laymen, recent motion generation methods can directly generate human motions conditioned on natural languages. However, it remains challenging to achieve diverse and fine-grained motion generation with various text inputs. To address this problem, we propose MotionDiffuse, the first diffusion model-based text-driven motion generation framework, which demonstrates several desired properties over existing methods. 1) Probabilistic Mapping. Instead of a deterministic language-motion mapping, MotionDiffuse generates motions through a series of denoising steps in which variations are injected. 2) Realistic Synthesis. MotionDiffuse excels at modeling complicated data distribution and generating vivid motion sequences. 3) Multi-Level Manipulation. MotionDiffuse responds to fine-grained instructions on body parts, and arbitrary-length motion synthesis with time-varied text prompts. Our experiments show MotionDiffuse outperforms existing SoTA methods by convincing margins on text-driven motion generation and action-conditioned motion generation. A qualitative analysis further demonstrates MotionDiffuse's controllability for comprehensive motion generation. Homepage: this https URL

Aug28 - Sep3, 2022

• Multi-View Reconstruction using Signed Ray Distance Functions (SRDF) | [code]

  In this paper, we address the problem of multi-view 3D shape reconstruction. While recent differentiable rendering approaches associated to implicit shape representations have provided breakthrough performance, they are still computationally heavy and often lack precision on the estimated geometries. To overcome these limitations we investigate a new computational approach that builds on a novel shape representation that is volumetric, as in recent differentiable rendering approaches, but parameterized with depth maps to better materialize the shape surface. The shape energy associated to this representation evaluates 3D geometry given color images and does not need appearance prediction but still benefits from volumetric integration when optimized. In practice we propose an implicit shape representation, the SRDF, based on signed distances which we parameterize by depths along camera rays. The associated shape energy considers the agreement between depth prediction consistency and photometric consistency, this at 3D locations within the volumetric representation. Various photo-consistency priors can be accounted for such as a median based baseline, or a more elaborated criterion as with a learned function. The approach retains pixel-accuracy with depth maps and is parallelizable. Our experiments over standard datasets shows that it provides state-of-the-art results with respect to recent approaches with implicit shape representations as well as with respect to traditional multi-view stereo methods.

• Dual-Space NeRF: Learning Animatable Avatars and Scene Lighting in Separate Spaces, 3DV2022 | [code]

  Modeling the human body in a canonical space is a common practice for capturing and animation. But when involving the neural radiance field (NeRF), learning a static NeRF in the canonical space is not enough because the lighting of the body changes when the person moves even though the scene lighting is constant. Previous methods alleviate the inconsistency of lighting by learning a per-frame embedding, but this operation does not generalize to unseen poses. Given that the lighting condition is static in the world space while the human body is consistent in the canonical space, we propose a dual-space NeRF that models the scene lighting and the human body with two MLPs in two separate spaces. To bridge these two spaces, previous methods mostly rely on the linear blend skinning (LBS) algorithm. However, the blending weights for LBS of a dynamic neural field are intractable and thus are usually memorized with another MLP, which does not generalize to novel poses. Although it is possible to borrow the blending weights of a parametric mesh such as SMPL, the interpolation operation introduces more artifacts. In this paper, we propose to use the barycentric mapping, which can directly generalize to unseen poses and surprisingly achieves superior results than LBS with neural blending weights. Quantitative and qualitative results on the Human3.6M and the ZJU-MoCap datasets show the effectiveness of our method.

• FoV-NeRF: Foveated Neural Radiance Fields for Virtual Reality, TVCG2022 | [code]

  Virtual Reality (VR) is becoming ubiquitous with the rise of consumer displays and commercial VR platforms. Such displays require low latency and high quality rendering of synthetic imagery with reduced compute overheads. Recent advances in neural rendering showed promise of unlocking new possibilities in 3D computer graphics via image-based representations of virtual or physical environments. Specifically, the neural radiance fields (NeRF) demonstrated that photo-realistic quality and continuous view changes of 3D scenes can be achieved without loss of view-dependent effects. While NeRF can significantly benefit rendering for VR applications, it faces unique challenges posed by high field-of-view, high resolution, and stereoscopic/egocentric viewing, typically causing low quality and high latency of the rendered images. In VR, this not only harms the interaction experience but may also cause sickness. To tackle these problems toward six-degrees-of-freedom, egocentric, and stereo NeRF in VR, we present the first gaze-contingent 3D neural representation and view synthesis method . We incorporate the human psychophysics of visual- and stereo-acuity into an egocentric neural representation of 3D scenery. We then jointly optimize the latency/performance and visual quality while mutually bridging human perception and neural scene synthesis to achieve perceptually high-quality immersive interaction. We conducted both objective analysis and subjective studies to evaluate the effectiveness of our approach. We find that our method significantly reduces latency (up to 99% time reduction compared with NeRF) without loss of high-fidelity rendering (perceptually identical to full-resolution ground truth). The presented approach may serve as the first step toward future VR/AR systems that capture, teleport, and visualize remote environments in real-time.

• Cross-Spectral Neural Radiance Fields, 3DV2022 | [code]

  We propose X-NeRF, a novel method to learn a Cross-Spectral scene representation given images captured from cameras with different light spectrum sensitivity, based on the Neural Radiance Fields formulation. X-NeRF optimizes camera poses across spectra during training and exploits Normalized Cross-Device Coordinates (NXDC) to render images of different modalities from arbitrary viewpoints, which are aligned and at the same resolution. Experiments on 16 forward-facing scenes, featuring color, multi-spectral and infrared images, confirm the effectiveness of X-NeRF at modeling Cross-Spectral scene representations.

• CLONeR: Camera-Lidar Fusion for Occupancy Grid-aided Neural Representations | [code]

  This paper proposes CLONeR, which significantly improves upon NeRF by allowing it to model large outdoor driving scenes that are observed from sparse input sensor views. This is achieved by decoupling occupancy and color learning within the NeRF framework into separate Multi-Layer Perceptrons (MLPs) trained using LiDAR and camera data, respectively. In addition, this paper proposes a novel method to build differentiable 3D Occupancy Grid Maps (OGM) alongside the NeRF model, and leverage this occupancy grid for improved sampling of points along a ray for volumetric rendering in metric space.

• NerfCap: Human Performance Capture With Dynamic Neural Radiance Fields, TVCG2022 | [code]

  This paper addresses the challenge of human performance capture from sparse multi-view or monocular videos. Given a template mesh of the performer, previous methods capture the human motion by non-rigidly registering the template mesh to images with 2D silhouettes or dense photometric alignment. However, the detailed surface deformation cannot be recovered from the silhouettes, while the photometric alignment suffers from instability caused by appearance variation in the videos. To solve these problems, we propose NerfCap, a novel performance capture method based on the dynamic neural radiance field (NeRF) representation of the performer. Specifically, a canonical NeRF is initialized from the template geometry and registered to the video frames by optimizing the deformation field and the appearance model of the canonical NeRF. To capture both large body motion and detailed surface deformation, NerfCap combines linear blend skinning with embedded graph deformation. In contrast to the mesh-based methods that suffer from fixed topology and texture, NerfCap is able to flexibly capture complex geometry and appearance variation across the videos, and synthesize more photo-realistic images. In addition, NerfCap can be pre-trained end to end in a self-supervised manner by matching the synthesized videos with the input videos. Experimental results on various datasets show that NerfCap outperforms prior works in terms of both surface reconstruction accuracy and novel-view synthesis quality.

Aug21 - Aug27, 2022

• Training and Tuning Generative Neural Radiance Fields for Attribute-Conditional 3D-Aware Face Generation | [code]

  3D-aware GANs based on generative neural radiance fields (GNeRF) have achieved impressive high-quality image generation, while preserving strong 3D consistency. The most notable achievements are made in the face generation domain. However, most of these models focus on improving view consistency but neglect a disentanglement aspect, thus these models cannot provide high-quality semantic/attribute control over generation. To this end, we introduce a conditional GNeRF model that uses specific attribute labels as input in order to improve the controllabilities and disentangling abilities of 3D-aware generative models. We utilize the pre-trained 3D-aware model as the basis and integrate a dual-branches attribute-editing module (DAEM), that utilize attribute labels to provide control over generation. Moreover, we propose a TRIOT (TRaining as Init, and Optimizing for Tuning) method to optimize the latent vector to improve the precision of the attribute-editing further. Extensive experiments on the widely used FFHQ show that our model yields high-quality editing with better view consistency while preserving the non-target regions. The code is available at this https URL.

• Voxurf: Voxel-based Efficient and Accurate Neural Surface Reconstruction | [code]

  Neural surface reconstruction aims to reconstruct accurate 3D surfaces based on multi-view images. Previous methods based on neural volume rendering mostly train a fully implicit model, and they require hours of training for a single scene. Recent efforts explore the explicit volumetric representation, which substantially accelerates the optimization process by memorizing significant information in learnable voxel grids. However, these voxel-based methods often struggle in reconstructing fine-grained geometry. Through empirical studies, we found that high-quality surface reconstruction hinges on two key factors: the capability of constructing a coherent shape and the precise modeling of color-geometry dependency. In particular, the latter is the key to the accurate reconstruction of fine details. Inspired by these findings, we develop Voxurf, a voxel-based approach for efficient and accurate neural surface reconstruction, which consists of two stages: 1) leverage a learnable feature grid to construct the color field and obtain a coherent coarse shape, and 2) refine detailed geometry with a dual color network that captures precise color-geometry dependency. We further introduce a hierarchical geometry feature to enable information sharing across voxels. Our experiments show that Voxurf achieves high efficiency and high quality at the same time. On the DTU benchmark, Voxurf achieves higher reconstruction quality compared to state-of-the-art methods, with 20x speedup in training.

• Neural Novel Actor: Learning a Generalized Animatable Neural Representation for Human Actors | [code]

  We propose a new method for learning a generalized animatable neural human representation from a sparse set of multi-view imagery of multiple persons. The learned representation can be used to synthesize novel view images of an arbitrary person from a sparse set of cameras, and further animate them with the user's pose control. While existing methods can either generalize to new persons or synthesize animations with user control, none of them can achieve both at the same time. We attribute this accomplishment to the employment of a 3D proxy for a shared multi-person human model, and further the warping of the spaces of different poses to a shared canonical pose space, in which we learn a neural field and predict the person- and pose-dependent deformations, as well as appearance with the features extracted from input images. To cope with the complexity of the large variations in body shapes, poses, and clothing deformations, we design our neural human model with disentangled geometry and appearance. Furthermore, we utilize the image features both at the spatial point and on the surface points of the 3D proxy for predicting person- and pose-dependent properties. Experiments show that our method significantly outperforms the state-of-the-arts on both tasks. The video and code are available at this https URL.

• DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation | [code]

  Large text-to-image models achieved a remarkable leap in the evolution of AI, enabling high-quality and diverse synthesis of images from a given text prompt. However, these models lack the ability to mimic the appearance of subjects in a given reference set and synthesize novel renditions of them in different contexts. In this work, we present a new approach for "personalization" of text-to-image diffusion models (specializing them to users' needs). Given as input just a few images of a subject, we fine-tune a pretrained text-to-image model (Imagen, although our method is not limited to a specific model) such that it learns to bind a unique identifier with that specific subject. Once the subject is embedded in the output domain of the model, the unique identifier can then be used to synthesize fully-novel photorealistic images of the subject contextualized in different scenes. By leveraging the semantic prior embedded in the model with a new autogenous class-specific prior preservation loss, our technique enables synthesizing the subject in diverse scenes, poses, views, and lighting conditions that do not appear in the reference images. We apply our technique to several previously-unassailable tasks, including subject recontextualization, text-guided view synthesis, appearance modification, and artistic rendering (all while preserving the subject's key features). Project page: this https URL

• E-NeRF: Neural Radiance Fields from a Moving Event Camera | [code]

  Estimating neural radiance fields (NeRFs) from ideal images has been extensively studied in the computer vision community. Most approaches assume optimal illumination and slow camera motion. These assumptions are often violated in robotic applications, where images contain motion blur and the scene may not have suitable illumination. This can cause significant problems for downstream tasks such as navigation, inspection or visualization of the scene. To alleviate these problems we present E-NeRF, the first method which estimates a volumetric scene representation in the form of a NeRF from a fast-moving event camera. Our method can recover NeRFs during very fast motion and in high dynamic range conditions, where frame-based approaches fail. We show that rendering high-quality frames is possible by only providing an event stream as input. Furthermore, by combining events and frames, we can estimate NeRFs of higher quality than state-of-the-art approaches under severe motion blur. We also show that combining events and frames can overcome failure cases of NeRF estimation in scenarios where only few input views are available, without requiring additional regularization.

• FurryGAN: High Quality Foreground-aware Image Synthesis, ECCV2022 | [code]

  Foreground-aware image synthesis aims to generate images as well as their foreground masks. A common approach is to formulate an image as an masked blending of a foreground image and a background image. It is a challenging problem because it is prone to reach the trivial solution where either image overwhelms the other, i.e., the masks become completely full or empty, and the foreground and background are not meaningfully separated. We present FurryGAN with three key components: 1) imposing both the foreground image and the composite image to be realistic, 2) designing a mask as a combination of coarse and fine masks, and 3) guiding the generator by an auxiliary mask predictor in the discriminator. Our method produces realistic images with remarkably detailed alpha masks which cover hair, fur, and whiskers in a fully unsupervised manner.

• SCONE: Surface Coverage Optimization in Unknown Environments by Volumetric Integration | [code]

  Next Best View computation (NBV) is a long-standing problem in robotics, and consists in identifying the next most informative sensor position(s) for reconstructing a 3D object or scene efficiently and accurately. Like most current methods, we consider NBV prediction from a depth sensor. Learning-based methods relying on a volumetric representation of the scene are suitable for path planning, but do not scale well with the size of the scene and have lower accuracy than methods using a surface-based representation. However, the latter constrain the camera to a small number of poses. To obtain the advantages of both representations, we show that we can maximize surface metrics by Monte Carlo integration over a volumetric representation. Our method scales to large scenes and handles free camera motion: It takes as input an arbitrarily large point cloud gathered by a depth sensor like Lidar systems as well as camera poses to predict NBV. We demonstrate our approach on a novel dataset made of large and complex 3D scenes.

Aug14 - Aug20, 2022

• Vox-Surf: Voxel-based Implicit Surface Representation | [code]

  Virtual content creation and interaction play an important role in modern 3D applications such as AR and VR. Recovering detailed 3D models from real scenes can significantly expand the scope of its applications and has been studied for decades in the computer vision and computer graphics community. We propose Vox-Surf, a voxel-based implicit surface representation. Our Vox-Surf divides the space into finite bounded voxels. Each voxel stores geometry and appearance information in its corner vertices. Vox-Surf is suitable for almost any scenario thanks to sparsity inherited from voxel representation and can be easily trained from multiple view images. We leverage the progressive training procedure to extract important voxels gradually for further optimization so that only valid voxels are preserved, which greatly reduces the number of sampling points and increases rendering speed.The fine voxels can also be considered as the bounding volume for collision detection.The experiments show that Vox-Surf representation can learn delicate surface details and accurate color with less memory and faster rendering speed than other methods.We also show that Vox-Surf can be more practical in scene editing and AR applications.

• Temporal View Synthesis of Dynamic Scenes through 3D Object Motion Estimation with Multi-Plane Images, ISMAR2022 | [code]

  The challenge of graphically rendering high frame-rate videos on low compute devices can be addressed through periodic prediction of future frames to enhance the user experience in virtual reality applications. This is studied through the problem of temporal view synthesis (TVS), where the goal is to predict the next frames of a video given the previous frames and the head poses of the previous and the next frames. In this work, we consider the TVS of dynamic scenes in which both the user and objects are moving. We design a framework that decouples the motion into user and object motion to effectively use the available user motion while predicting the next frames. We predict the motion of objects by isolating and estimating the 3D object motion in the past frames and then extrapolating it. We employ multi-plane images (MPI) as a 3D representation of the scenes and model the object motion as the 3D displacement between the corresponding points in the MPI representation. In order to handle the sparsity in MPIs while estimating the motion, we incorporate partial convolutions and masked correlation layers to estimate corresponding points. The predicted object motion is then integrated with the given user or camera motion to generate the next frame. Using a disocclusion infilling module, we synthesize the regions uncovered due to the camera and object motion. We develop a new synthetic dataset for TVS of dynamic scenes consisting of 800 videos at full HD resolution. We show through experiments on our dataset and the MPI Sintel dataset that our model outperforms all the competing methods in the literature.

• LoRD: Local 4D Implicit Representation for High-Fidelity Dynamic Human Modeling, ECCV2022 | [code]

  Recent progress in 4D implicit representation focuses on globally controlling the shape and motion with low dimensional latent vectors, which is prone to missing surface details and accumulating tracking error. While many deep local representations have shown promising results for 3D shape modeling, their 4D counterpart does not exist yet. In this paper, we fill this blank by proposing a novel Local 4D implicit Representation for Dynamic clothed human, named LoRD, which has the merits of both 4D human modeling and local representation, and enables high-fidelity reconstruction with detailed surface deformations, such as clothing wrinkles. Particularly, our key insight is to encourage the network to learn the latent codes of local part-level representation, capable of explaining the local geometry and temporal deformations. To make the inference at test-time, we first estimate the inner body skeleton motion to track local parts at each time step, and then optimize the latent codes for each part via auto-decoding based on different types of observed data. Extensive experiments demonstrate that the proposed method has strong capability for representing 4D human, and outperforms state-of-the-art methods on practical applications, including 4D reconstruction from sparse points, non-rigid depth fusion, both qualitatively and quantitatively.

• Neural Capture of Animatable 3D Human from Monocular Video, ECCV2022 | [code]

  We present a novel paradigm of building an animatable 3D human representation from a monocular video input, such that it can be rendered in any unseen poses and views. Our method is based on a dynamic Neural Radiance Field (NeRF) rigged by a mesh-based parametric 3D human model serving as a geometry proxy. Previous methods usually rely on multi-view videos or accurate 3D geometry information as additional inputs; besides, most methods suffer from degraded quality when generalized to unseen poses. We identify that the key to generalization is a good input embedding for querying dynamic NeRF: A good input embedding should define an injective mapping in the full volumetric space, guided by surface mesh deformation under pose variation. Based on this observation, we propose to embed the input query with its relationship to local surface regions spanned by a set of geodesic nearest neighbors on mesh vertices. By including both position and relative distance information, our embedding defines a distance-preserved deformation mapping and generalizes well to unseen poses. To reduce the dependency on additional inputs, we first initialize per-frame 3D meshes using off-the-shelf tools and then propose a pipeline to jointly optimize NeRF and refine the initial mesh. Extensive experiments show our method can synthesize plausible human rendering results under unseen poses and views.

• The 8-Point Algorithm as an Inductive Bias for Relative Pose Prediction by ViTs, 3DV2022 | [code]

  We present a simple baseline for directly estimating the relative pose (rotation and translation, including scale) between two images. Deep methods have recently shown strong progress but often require complex or multi-stage architectures. We show that a handful of modifications can be applied to a Vision Transformer (ViT) to bring its computations close to the Eight-Point Algorithm. This inductive bias enables a simple method to be competitive in multiple settings, often substantially improving over the state of the art with strong performance gains in limited data regimes.

• PDRF: Progressively Deblurring Radiance Field for Fast and Robust Scene Reconstruction from Blurry Images | [code]

  We present Progressively Deblurring Radiance Field (PDRF), a novel approach to efficiently reconstruct high quality radiance fields from blurry images. While current State-of-The-Art (SoTA) scene reconstruction methods achieve photo-realistic rendering results from clean source views, their performances suffer when the source views are affected by blur, which is commonly observed for images in the wild. Previous deblurring methods either do not account for 3D geometry, or are computationally intense. To addresses these issues, PDRF, a progressively deblurring scheme in radiance field modeling, accurately models blur by incorporating 3D scene context. PDRF further uses an efficient importance sampling scheme, which results in fast scene optimization. Specifically, PDRF proposes a Coarse Ray Renderer to quickly estimate voxel density and feature; a Fine Voxel Renderer is then used to achieve high quality ray tracing. We perform extensive experiments and show that PDRF is 15X faster than previous SoTA while achieving better performance on both synthetic and real scenes.

• Text-to-Image Generation via Implicit Visual Guidance and Hypernetwork | [code]

  We develop an approach for text-to-image generation that embraces additional retrieval images, driven by a combination of implicit visual guidance loss and generative objectives. Unlike most existing text-to-image generation methods which merely take the text as input, our method dynamically feeds cross-modal search results into a unified training stage, hence improving the quality, controllability and diversity of generation results. We propose a novel hypernetwork modulated visual-text encoding scheme to predict the weight update of the encoding layer, enabling effective transfer from visual information (e.g. layout, content) into the corresponding latent domain. Experimental results show that our model guided with additional retrieval visual data outperforms existing GAN-based models. On COCO dataset, we achieve better FID of 9.13 with up to 3.5× fewer generator parameters, compared with the state-of-the-art method.

• Casual Indoor HDR Radiance Capture from Omnidirectional Images | [code]

  We present PanoHDR-NeRF, a novel pipeline to casually capture a plausible full HDR radiance field of a large indoor scene without elaborate setups or complex capture protocols. First, a user captures a low dynamic range (LDR) omnidirectional video of the scene by freely waving an off-the-shelf camera around the scene. Then, an LDR2HDR network uplifts the captured LDR frames to HDR, subsequently used to train a tailored NeRF++ model. The resulting PanoHDR-NeRF pipeline can estimate full HDR panoramas from any location of the scene. Through experiments on a novel test dataset of a variety of real scenes with the ground truth HDR radiance captured at locations not seen during training, we show that PanoHDR-NeRF predicts plausible radiance from any scene point. We also show that the HDR images produced by PanoHDR-NeRF can synthesize correct lighting effects, enabling the augmentation of indoor scenes with synthetic objects that are lit correctly.

• UPST-NeRF: Universal Photorealistic Style Transfer of Neural Radiance Fields for 3D Scene | [code]

  3D scenes photorealistic stylization aims to generate photorealistic images from arbitrary novel views according to a given style image while ensuring consistency when rendering from different viewpoints. Some existing stylization methods with neural radiance fields can effectively predict stylized scenes by combining the features of the style image with multi-view images to train 3D scenes. However, these methods generate novel view images that contain objectionable artifacts. Besides, they cannot achieve universal photorealistic stylization for a 3D scene. Therefore, a styling image must retrain a 3D scene representation network based on a neural radiation field. We propose a novel 3D scene photorealistic style transfer framework to address these issues. It can realize photorealistic 3D scene style transfer with a 2D style image. We first pre-trained a 2D photorealistic style transfer network, which can meet the photorealistic style transfer between any given content image and style image. Then, we use voxel features to optimize a 3D scene and get the geometric representation of the scene. Finally, we jointly optimize a hyper network to realize the scene photorealistic style transfer of arbitrary style images. In the transfer stage, we use a pre-trained 2D photorealistic network to constrain the photorealistic style of different views and different style images in the 3D scene. The experimental results show that our method not only realizes the 3D photorealistic style transfer of arbitrary style images but also outperforms the existing methods in terms of visual quality and consistency. Project page:this https URL.

• DM-NeRF: 3D Scene Geometry Decomposition and Manipulation from 2D Images | [code]

  In this paper, we study the problem of 3D scene geometry decomposition and manipulation from 2D views. By leveraging the recent implicit neural representation techniques, particularly the appealing neural radiance fields, we introduce an object field component to learn unique codes for all individual objects in 3D space only from 2D supervision. The key to this component is a series of carefully designed loss functions to enable every 3D point, especially in non-occupied space, to be effectively optimized even without 3D labels. In addition, we introduce an inverse query algorithm to freely manipulate any specified 3D object shape in the learned scene representation. Notably, our manipulation algorithm can explicitly tackle key issues such as object collisions and visual occlusions. Our method, called DM-NeRF, is among the first to simultaneously reconstruct, decompose, manipulate and render complex 3D scenes in a single pipeline. Extensive experiments on three datasets clearly show that our method can accurately decompose all 3D objects from 2D views, allowing any interested object to be freely manipulated in 3D space such as translation, rotation, size adjustment, and deformation.

• HDR-Plenoxels: Self-Calibrating High Dynamic Range Radiance Fields, ECCV2022 | [code]

  We propose high dynamic range radiance (HDR) fields, HDR-Plenoxels, that learn a plenoptic function of 3D HDR radiance fields, geometry information, and varying camera settings inherent in 2D low dynamic range (LDR) images. Our voxel-based volume rendering pipeline reconstructs HDR radiance fields with only multi-view LDR images taken from varying camera settings in an end-to-end manner and has a fast convergence speed. To deal with various cameras in real-world scenarios, we introduce a tone mapping module that models the digital in-camera imaging pipeline (ISP) and disentangles radiometric settings. Our tone mapping module allows us to render by controlling the radiometric settings of each novel view. Finally, we build a multi-view dataset with varying camera conditions, which fits our problem setting. Our experiments show that HDR-Plenoxels can express detail and high-quality HDR novel views from only LDR images with various cameras.

Aug7 - Aug13, 2022

• Progressive Multi-scale Light Field Networks, 3DV2022 | [code]

  Neural representations have shown great promise in their ability to represent radiance and light fields while being very compact compared to the image set representation. However, current representations are not well suited for streaming as decoding can only be done at a single level of detail and requires downloading the entire neural network model. Furthermore, high-resolution light field networks can exhibit flickering and aliasing as neural networks are sampled without appropriate filtering. To resolve these issues, we present a progressive multi-scale light field network that encodes a light field with multiple levels of detail. Lower levels of detail are encoded using fewer neural network weights enabling progressive streaming and reducing rendering time. Our progressive multi-scale light field network addresses aliasing by encoding smaller anti-aliased representations at its lower levels of detail. Additionally, per-pixel level of detail enables our representation to support dithered transitions and foveated rendering.

• OmniVoxel: A Fast and Precise Reconstruction Method of Omnidirectional Neural Radiance Field, GCCE 2022 | [code]

  This paper proposes a method to reconstruct the neural radiance field with equirectangular omnidirectional images. Implicit neural scene representation with a radiance field can reconstruct the 3D shape of a scene continuously within a limited spatial area. However, training a fully implicit representation on commercial PC hardware requires a lot of time and computing resources (15 ∼ 20 hours per scene). Therefore, we propose a method to accelerate this process significantly (20 ∼ 40 minutes per scene). Instead of using a fully implicit representation of rays for radiance field reconstruction, we adopt feature voxels that contain density and color features in tensors. Considering omnidirectional equirectangular input and the camera layout, we use spherical voxelization for representation instead of cubic representation. Our voxelization method could balance the reconstruction quality of the inner scene and outer scene. In addition, we adopt the axis-aligned positional encoding method on the color features to increase the total image quality. Our method achieves satisfying empirical performance on synthetic datasets with random camera poses. Moreover, we test our method with real scenes which contain complex geometries and also achieve state-of-the-art performance. Our code and complete dataset will be released at the same time as the paper publication.

• HyperTime: Implicit Neural Representation for Time Series | [code]

  Implicit neural representations (INRs) have recently emerged as a powerful tool that provides an accurate and resolution-independent encoding of data. Their robustness as general approximators has been shown in a wide variety of data sources, with applications on image, sound, and 3D scene representation. However, little attention has been given to leveraging these architectures for the representation and analysis of time series data. In this paper, we analyze the representation of time series using INRs, comparing different activation functions in terms of reconstruction accuracy and training convergence speed. We show how these networks can be leveraged for the imputation of time series, with applications on both univariate and multivariate data. Finally, we propose a hypernetwork architecture that leverages INRs to learn a compressed latent representation of an entire time series dataset. We introduce an FFT-based loss to guide training so that all frequencies are preserved in the time series. We show that this network can be used to encode time series as INRs, and their embeddings can be interpolated to generate new time series from existing ones. We evaluate our generative method by using it for data augmentation, and show that it is competitive against current state-of-the-art approaches for augmentation of time series.

• RelPose: Predicting Probabilistic Relative Rotation for Single Objects in the Wild, ECCV2022 | [code]

  We describe a data-driven method for inferring the camera viewpoints given multiple images of an arbitrary object. This task is a core component of classic geometric pipelines such as SfM and SLAM, and also serves as a vital pre-processing requirement for contemporary neural approaches (e.g. NeRF) to object reconstruction and view synthesis. In contrast to existing correspondence-driven methods that do not perform well given sparse views, we propose a top-down prediction based approach for estimating camera viewpoints. Our key technical insight is the use of an energy-based formulation for representing distributions over relative camera rotations, thus allowing us to explicitly represent multiple camera modes arising from object symmetries or views. Leveraging these relative predictions, we jointly estimate a consistent set of camera rotations from multiple images. We show that our approach outperforms state-of-the-art SfM and SLAM methods given sparse images on both seen and unseen categories. Further, our probabilistic approach significantly outperforms directly regressing relative poses, suggesting that modeling multimodality is important for coherent joint reconstruction. We demonstrate that our system can be a stepping stone toward in-the-wild reconstruction from multi-view datasets. The project page with code and videos can be found at this https URL.

• NIDN: Neural Inverse Design of Nanostructures | [code]

  In the recent decade, computational tools have become central in material design, allowing rapid development cycles at reduced costs. Machine learning tools are especially on the rise in photonics. However, the inversion of the Maxwell equations needed for the design is particularly challenging from an optimization standpoint, requiring sophisticated software. We present an innovative, open-source software tool called Neural Inverse Design of Nanostructures (NIDN) that allows designing complex, stacked material nanostructures using a physics-based deep learning approach. Instead of a derivative-free or data-driven optimization or learning method, we perform a gradient-based neural network training where we directly optimize the material and its structure based on its spectral characteristics. NIDN supports two different solvers, rigorous coupled-wave analysis and a finite-difference time-domain method. The utility and validity of NIDN are demonstrated on several synthetic examples as well as the design of a 1550 nm filter and anti-reflection coating. Results match experimental baselines, other simulation tools, and the desired spectral characteristics. Given its full modularity in regard to network architectures and Maxwell solvers as well as open-source, permissive availability, NIDN will be able to support computational material design processes in a broad range of applications.

• HRF-Net: Holistic Radiance Fields from Sparse Inputs | [code]

  We present HRF-Net, a novel view synthesis method based on holistic radiance fields that renders novel views using a set of sparse inputs. Recent generalizing view synthesis methods also leverage the radiance fields but the rendering speed is not real-time. There are existing methods that can train and render novel views efficiently but they can not generalize to unseen scenes. Our approach addresses the problem of real-time rendering for generalizing view synthesis and consists of two main stages: a holistic radiance fields predictor and a convolutional-based neural renderer. This architecture infers not only consistent scene geometry based on the implicit neural fields but also renders new views efficiently using a single GPU. We first train HRF-Net on multiple 3D scenes of the DTU dataset and the network can produce plausible novel views on unseen real and synthetics data using only photometric losses. Moreover, our method can leverage a denser set of reference images of a single scene to produce accurate novel views without relying on additional explicit representations and still maintains the high-speed rendering of the pre-trained model. Experimental results show that HRF-Net outperforms state-of-the-art generalizable neural rendering methods on various synthetic and real datasets.

• Monte Carlo Denoising Using Implicit Neural Representation | [code]

  Monte Carlo path tracing is a popular 3D rendering technique in computer graphics, but it often requires a costly tradeoff between the amount of noise in the image and computation time. Therefore, it is useful to attempt to “smooth out” a noisy image, typically by constructing new data between the samples or applying filters to the image. In this work, we investigate the feasibility of training a neural network to implicitly represent the radiance of a fixed-viewpoint scene as a continuous function. We implement the neural network using a multilayer perceptron network and train it on a sparsely sampled image that is generated by an offline Monte Carlo renderer. This training data uses the (x, y) coordinate of each sample on the image plane as inputs and the RGB color of the sample as outputs. Additionally, we provide the network with the surface normal, depth, and albedo of the first ray intersection as extra inputs alongside the pixel coordinates. These extra input dimensions improve the quality of the implicit representation by helping the network account for changes in depth, normal, and diffuse color. Once the network is trained on the sparsely sampled scene, we can densely sample the network many times per pixel to create the final denoised image. We find that this network can quickly learn and denoise images in scenes with soft lighting and glossy reflections, and it can easily handle discontinuities in depth, normal, and diffuse color with just a small amount of training.

• Fast Gradient Descent for Surface Capture Via Differentiable Rendering, 3DV2022 | [code]

  Differential rendering has recently emerged as a powerful tool for image-based rendering or geometric reconstruction from multiple views, with very high quality. Up to now, such methods have been benchmarked on generic object databases and promisingly applied to some real data, but have yet to be applied to specific applications that may benefit. In this paper, we investigate how a differential rendering system can be crafted for raw multi-camera performance capture. We address several key issues in the way of practical usability and reproducibility, such as processing speed, explainability of the model, and general output model quality. This leads us to several contributions to the differential rendering framework. In particular we show that a unified view of differential rendering and classic optimization is possible, leading to a formulation and implementation where complete non-stochastic gradient steps can be analytically computed and the full perframe data stored in video memory, yielding a straightforward and efficient implementation. We also use a sparse storage and coarse-to-fine scheme to achieve extremely high resolution with contained memory and computation time. We show experimentally that results rivaling in quality with state of the art multi-view human surface capture methods are achievable in a fraction of the time, typically around a minute per frame.

• PlaneFormers: From Sparse View Planes to 3D Reconstruction, ECCV2022 | [code]

  We present an approach for the planar surface reconstruction of a scene from images with limited overlap. This reconstruction task is challenging since it requires jointly reasoning about single image 3D reconstruction, correspondence between images, and the relative camera pose between images. Past work has proposed optimization-based approaches. We introduce a simpler approach, the PlaneFormer, that uses a transformer applied to 3D-aware plane tokens to perform 3D reasoning. Our experiments show that our approach is substantially more effective than prior work, and that several 3D-specific design decisions are crucial for its success.

• PS-NeRV: Patch-wise Stylized Neural Representations for Videos | [code]

  We study how to represent a video with implicit neural representations (INRs). Classical INRs methods generally utilize MLPs to map input coordinates to output pixels. While some recent works have tried to directly reconstruct the whole image with CNNs. However, we argue that both the above pixel-wise and image-wise strategies are not favorable to video data. Instead, we propose a patch-wise solution, PS-NeRV, which represents videos as a function of patches and the corresponding patch coordinate. It naturally inherits the advantages of image-wise methods, and achieves excellent reconstruction performance with fast decoding speed. The whole method includes conventional modules, like positional embedding, MLPs and CNNs, while also introduces AdaIN to enhance intermediate features. These simple yet essential changes could help the network easily fit high-frequency details. Extensive experiments have demonstrated its effectiveness in several video-related tasks, such as video compression and video inpainting.

Jul31 - Aug6, 2022

• NFOMP: Neural Field for Optimal Motion Planner of Differential Drive Robots With Nonholonomic Constraints, IEEE Robotics and Automation Letters, IEEE Robotics and Automation Letters | [code]

  Optimal motion planning is one of the most critical problems in mobile robotics. On the one hand, classical sampling-based methods propose asymptotically optimal solutions to this problem. However, these planners cannot achieve smooth and short trajectories in reasonable calculation time. On the other hand, optimization-based methods are able to generate smooth and plain trajectories in a variety of scenarios, including a dense human crowd. However, modern optimization-based methods use the precomputed signed distance function for collision loss estimation, and it limits the application of these methods for general configuration spaces, including a differential drive non-circular robot with non-holonomic constraints. Moreover, optimization-based methods lack the ability to handle U-shaped or thin obstacles accurately. We propose to improve the optimization methods in two aspects. Firstly, we developed an obstacle neural field model to estimate collision loss; training this model together with trajectory optimization allows improving collision loss continuously, while achieving more feasible and smoother trajectories. Secondly, we forced the trajectory to consider non-holonomic constraints by adding Lagrange multipliers to the trajectory loss function. We applied our method for solving the optimal motion planning problem for differential drive robots with non-holonomic constraints, benchmarked our solution, and proved that the novel planner generates smooth, short, and plain trajectories perfectly suitable for a robot to follow, and outperforms the state-of-the-art approaches by 25% on normalized curvature and by 75% on the number of cusps in the MovingAI environment.

• NeSF: Neural Semantic Fields for Generalizable Semantic Segmentation of 3D Scenes | [code]

  We present NeSF, a method for producing 3D semantic fields from pre-trained density fields and sparse 2D semantic supervision. Our method side-steps traditional scene representations by leveraging neural representations where 3D information is stored within neural fields. In spite of being supervised by 2D signals alone, our method is able to generate 3D-consistent semantic maps from novel camera poses and can be queried at arbitrary 3D points. Notably, NeSF is compatible with any method producing a density field, and its accuracy improves as the quality of the pre-trained density fields improve. Our empirical analysis demonstrates comparable quality to competitive 2D and 3D semantic segmentation baselines on convincing synthetic scenes while also offering features unavailable to existing methods.

• PRIF: Primary Ray-based Implicit Function | [code]

  We introduce a new implicit shape representation called Primary Ray-based Implicit Function (PRIF). In contrast to most existing approaches based on the signed distance function (SDF) which handles spatial locations, our representation operates on oriented rays. Specifically, PRIF is formulated to directly produce the surface hit point of a given input ray, without the expensive sphere-tracing operations, hence enabling efficient shape extraction and differentiable rendering. We demonstrate that neural networks trained to encode PRIF achieve successes in various tasks including single shape representation, category-wise shape generation, shape completion from sparse or noisy observations, inverse rendering for camera pose estimation, and neural rendering with color.

• Transformers as Meta-Learners for Implicit Neural Representations, ECCV2022 | [code]

  Implicit Neural Representations (INRs) have emerged and shown their benefits over discrete representations in recent years. However, fitting an INR to the given observations usually requires optimization with gradient descent from scratch, which is inefficient and does not generalize well with sparse observations. To address this problem, most of the prior works train a hypernetwork that generates a single vector to modulate the INR weights, where the single vector becomes an information bottleneck that limits the reconstruction precision of the output INR. Recent work shows that the whole set of weights in INR can be precisely inferred without the single-vector bottleneck by gradient-based meta-learning. Motivated by a generalized formulation of gradient-based meta-learning, we propose a formulation that uses Transformers as hypernetworks for INRs, where it can directly build the whole set of INR weights with Transformers specialized as set-to-set mapping. We demonstrate the effectiveness of our method for building INRs in different tasks and domains, including 2D image regression and view synthesis for 3D objects. Our work draws connections between the Transformer hypernetworks and gradient-based meta-learning algorithms and we provide further analysis for understanding the generated INRs.

• End-to-end learning of 3D phase-only holograms for holographic display | [code]

  Computer-generated holography (CGH) provides volumetric control of coherent wavefront and is fundamental to applications such as volumetric 3D displays, lithography, neural photostimulation, and optical/acoustic trapping. Recently, deep learning-based methods emerged as promising computational paradigms for CGH synthesis that overcome the quality-runtime tradeoff in conventional simulation/optimization-based methods. Yet, the quality of the predicted hologram is intrinsically bounded by the dataset’s quality. Here we introduce a new hologram dataset, MIT-CGH-4K-V2, that uses a layered depth image as a data-efficient volumetric 3D input and a two-stage supervised+unsupervised training protocol for direct synthesis of high-quality 3D phase-only holograms. The proposed system also corrects vision aberration, allowing customization for end-users. We experimentally show photorealistic 3D holographic projections and discuss relevant spatial light modulator calibration procedures. Our method runs in real-time on a consumer GPU and 5 FPS on an iPhone 13 Pro, promising drastically enhanced performance for the applications above.

• VolTeMorph: Realtime, Controllable and Generalisable Animation of Volumetric Representations | [code]

  The recent increase in popularity of volumetric representations for scene reconstruction and novel view synthesis has put renewed focus on animating volumetric content at high visual quality and in real-time. While implicit deformation methods based on learned functions can produce impressive results, they are `black boxes' to artists and content creators, they require large amounts of training data to generalise meaningfully, and they do not produce realistic extrapolations outside the training data. In this work we solve these issues by introducing a volume deformation method which is real-time, easy to edit with off-the-shelf software and can extrapolate convincingly. To demonstrate the versatility of our method, we apply it in two scenarios: physics-based object deformation and telepresence where avatars are controlled using blendshapes. We also perform thorough experiments showing that our method compares favourably to both volumetric approaches combined with implicit deformation and methods based on mesh deformation.

• Controllable Free Viewpoint Video Reconstruction Based on Neural Radiance Fields and Motion Graphs, IEEE Transactions on Visualization and Computer Graphics | [code]

  In this paper, we propose a controllable high-quality free viewpoint video generation method based on the motion graph and neural radiance fields (NeRF). Different from existing pose-driven NeRF or time/structure conditioned NeRF works, we propose to first construct a directed motion graph of the captured sequence. Such a sequence-motion-parameterization strategy not only enables flexible pose control for free viewpoint video rendering but also avoids redundant calculation of similar poses and thus improves the overall reconstruction efficiency. Moreover, to support body shape control without losing the realistic free viewpoint rendering performance, we improve the vanilla NeRF by combining explicit surface deformation and implicit neural scene representations. Specifically, we train a local surface-guided NeRF for each valid frame on the motion graph, and the volumetric rendering was only performed in the local space around the real surface, thus enabling plausible shape control ability. As far as we know, our method is the first method that supports both realistic free viewpoint video reconstruction and motion graph-based user-guided motion traversal. The results and comparisons further demonstrate the effectiveness of the proposed method.

• Robust Change Detection Based on Neural Descriptor Fields, IROS2022 | [code]

  The ability to reason about changes in the environment is crucial for robots operating over extended periods of time. Agents are expected to capture changes during operation so that actions can be followed to ensure a smooth progression of the working session. However, varying viewing angles and accumulated localization errors make it easy for robots to falsely detect changes in the surrounding world due to low observation overlap and drifted object associations. In this paper, based on the recently proposed category-level Neural Descriptor Fields (NDFs), we develop an object-level online change detection approach that is robust to partially overlapping observations and noisy localization results. Utilizing the shape completion capability and SE(3)-equivariance of NDFs, we represent objects with compact shape codes encoding full object shapes from partial observations. The objects are then organized in a spatial tree structure based on object centers recovered from NDFs for fast queries of object neighborhoods. By associating objects via shape code similarity and comparing local object-neighbor spatial layout, our proposed approach demonstrates robustness to low observation overlap and localization noises. We conduct experiments on both synthetic and real-world sequences and achieve improved change detection results compared to multiple baseline methods. Project webpage: this https URL

Jul24 - Jul30, 2022

• DoF-NeRF: Depth-of-Field Meets Neural Radiance Fields, ACMMM2022 | [code]

  Neural Radiance Field (NeRF) and its variants have exhibited great success on representing 3D scenes and synthesizing photo-realistic novel views. However, they are generally based on the pinhole camera model and assume all-in-focus inputs. This limits their applicability as images captured from the real world often have finite depth-of-field (DoF). To mitigate this issue, we introduce DoF-NeRF, a novel neural rendering approach that can deal with shallow DoF inputs and can simulate DoF effect. In particular, it extends NeRF to simulate the aperture of lens following the principles of geometric optics. Such a physical guarantee allows DoF-NeRF to operate views with different focus configurations. Benefiting from explicit aperture modeling, DoF-NeRF also enables direct manipulation of DoF effect by adjusting virtual aperture and focus parameters. It is plug-and-play and can be inserted into NeRF-based frameworks. Experiments on synthetic and real-world datasets show that, DoF-NeRF not only performs comparably with NeRF in the all-in-focus setting, but also can synthesize all-in-focus novel views conditioned on shallow DoF inputs. An interesting application of DoF-NeRF to DoF rendering is also demonstrated.

• ZEPI-Net: Light Field Super Resolution via Internal Cross-Scale Epipolar Plane Image Zero-Shot Learning, Neural Processing Letters (2022) | [code]

  Many applications of light field (LF) imaging have been limited by the spatial-angular resolution problem, hence the need for efficient super-resolution techniques. Recently, learning-based solutions have achieved remarkably better performances than traditional super-resolution (SR) techniques. Unfortunately, the learning or training process relies heavily on the training dataset, which could be limited for most LF imaging applications. In this paper, we propose a novel LF spatial-angular SR algorithm based on zero-shot learning. We suggest learning cross-scale reusable features in the epipolar plane image (EPI) space, and avoiding explicitly modeling scene priors or implicitly learning that from a large number of LFs. Most importantly, without using any external LFs, the proposed algorithm can simultaneously super-resolve a LF in both spatial and angular domains. Moreover, the proposed solution is free of depth or disparity estimation, which is usually employed by existing LF spatial and angular SR. By using a simple 8-layers fully convolutional network, we show that the proposed algorithm can generate comparable results to the state-of-the-art spatial SR. Our algorithm outperforms the existing methods in terms of angular SR on multiple groups of public LF datasets. The experiment results indicate that the cross-scale features can be well learned and be reused for LF SR in the EPI space.

• ObjectFusion: Accurate object-level SLAM with neural object priors, Graphical Models, Volume 123, September 2022 | [code]

  Previous object-level Simultaneous Localization and Mapping (SLAM) approaches still fail to create high quality object-oriented 3D map in an efficient way. The main challenges come from how to represent the object shape effectively and how to apply such object representation to accurate online camera tracking efficiently. In this paper, we provide ObjectFusion as a novel object-level SLAM in static scenes which efficiently creates object-oriented 3D map with high-quality object reconstruction, by leveraging neural object priors. We propose a neural object representation with only a single encoder–decoder network to effectively express the object shape across various categories, which benefits high quality reconstruction of object instance. More importantly, we propose to convert such neural object representation as precise measurements to jointly optimize the object shape, object pose and camera pose for the final accurate 3D object reconstruction. With extensive evaluations on synthetic and real-world RGB-D datasets, we show that our ObjectFusion outperforms previous approaches, with better object reconstruction quality, using much less memory footprint, and in a more efficient way, especially at the object level.

• MobileNeRF: Exploiting the Polygon Rasterization Pipeline for Efficient Neural Field Rendering on Mobile Architectures | [code]

  Neural Radiance Fields (NeRFs) have demonstrated amazing ability to synthesize images of 3D scenes from novel views. However, they rely upon specialized volumetric rendering algorithms based on ray marching that are mismatched to the capabilities of widely deployed graphics hardware. This paper introduces a new NeRF representation based on textured polygons that can synthesize novel images efficiently with standard rendering pipelines. The NeRF is represented as a set of polygons with textures representing binary opacities and feature vectors. Traditional rendering of the polygons with a z-buffer yields an image with features at every pixel, which are interpreted by a small, view-dependent MLP running in a fragment shader to produce a final pixel color. This approach enables NeRFs to be rendered with the traditional polygon rasterization pipeline, which provides massive pixel-level parallelism, achieving interactive frame rates on a wide range of compute platforms, including mobile phones.

• Neural Density-Distance Fields, ECCV2022 | [code]

  The success of neural fields for 3D vision tasks is now indisputable. Following this trend, several methods aiming for visual localization (e.g., SLAM) have been proposed to estimate distance or density fields using neural fields. However, it is difficult to achieve high localization performance by only density fields-based methods such as Neural Radiance Field (NeRF) since they do not provide density gradient in most empty regions. On the other hand, distance field-based methods such as Neural Implicit Surface (NeuS) have limitations in objects' surface shapes. This paper proposes Neural Density-Distance Field (NeDDF), a novel 3D representation that reciprocally constrains the distance and density fields. We extend distance field formulation to shapes with no explicit boundary surface, such as fur or smoke, which enable explicit conversion from distance field to density field. Consistent distance and density fields realized by explicit conversion enable both robustness to initial values and high-quality registration. Furthermore, the consistency between fields allows fast convergence from sparse point clouds. Experiments show that NeDDF can achieve high localization performance while providing comparable results to NeRF on novel view synthesis. The code is available at this https URL.

• End-to-end View Synthesis via NeRF Attention | [code]

  In this paper, we present a simple seq2seq formulation for view synthesis where we take a set of ray points as input and output colors corresponding to the rays. Directly applying a standard transformer on this seq2seq formulation has two limitations. First, the standard attention cannot successfully fit the volumetric rendering procedure, and therefore high-frequency components are missing in the synthesized views. Second, applying global attention to all rays and pixels is extremely inefficient. Inspired by the neural radiance field (NeRF), we propose the NeRF attention (NeRFA) to address the above problems. On the one hand, NeRFA considers the volumetric rendering equation as a soft feature modulation procedure. In this way, the feature modulation enhances the transformers with the NeRF-like inductive bias. On the other hand, NeRFA performs multi-stage attention to reduce the computational overhead. Furthermore, the NeRFA model adopts the ray and pixel transformers to learn the interactions between rays and pixels. NeRFA demonstrates superior performance over NeRF and NerFormer on four datasets: DeepVoxels, Blender, LLFF, and CO3D. Besides, NeRFA establishes a new state-of-the-art under two settings: the single-scene view synthesis and the category-centric novel view synthesis. The code will be made publicly available.

• Going Off-Grid: Continuous Implicit Neural Representations for 3D Vascular Modeling, MICCAI STACOM 2022 | [code]

  Personalised 3D vascular models are valuable for diagnosis, prognosis and treatment planning in patients with cardiovascular disease. Traditionally, such models have been constructed with explicit representations such as meshes and voxel masks, or implicit representations such as radial basis functions or atomic (tubular) shapes. Here, we propose to represent surfaces by the zero level set of their signed distance function (SDF) in a differentiable implicit neural representation (INR). This allows us to model complex vascular structures with a representation that is implicit, continuous, light-weight, and easy to integrate with deep learning algorithms. We here demonstrate the potential of this approach with three practical examples. First, we obtain an accurate and watertight surface for an abdominal aortic aneurysm (AAA) from CT images and show robust fitting from as little as 200 points on the surface. Second, we simultaneously fit nested vessel walls in a single INR without intersections. Third, we show how 3D models of individual arteries can be smoothly blended into a single watertight surface. Our results show that INRs are a flexible representation with potential for minimally interactive annotation and manipulation of complex vascular structures.

• Neural Strands: Learning Hair Geometry and Appearance from Multi-View Images, ECCV2022 | [code]

  We present Neural Strands, a novel learning framework for modeling accurate hair geometry and appearance from multi-view image inputs. The learned hair model can be rendered in real-time from any viewpoint with high-fidelity view-dependent effects. Our model achieves intuitive shape and style control unlike volumetric counterparts. To enable these properties, we propose a novel hair representation based on a neural scalp texture that encodes the geometry and appearance of individual strands at each texel location. Furthermore, we introduce a novel neural rendering framework based on rasterization of the learned hair strands. Our neural rendering is strand-accurate and anti-aliased, making the rendering view-consistent and photorealistic. Combining appearance with a multi-view geometric prior, we enable, for the first time, the joint learning of appearance and explicit hair geometry from a multi-view setup. We demonstrate the efficacy of our approach in terms of fidelity and efficiency for various hairstyles.

• Neural Radiance Transfer Fields for Relightable Novel-view Synthesis with Global Illumination | [code]

  Given a set of images of a scene, the re-rendering of this scene from novel views and lighting conditions is an important and challenging problem in Computer Vision and Graphics. On the one hand, most existing works in Computer Vision usually impose many assumptions regarding the image formation process, e.g. direct illumination and predefined materials, to make scene parameter estimation tractable. On the other hand, mature Computer Graphics tools allow modeling of complex photo-realistic light transport given all the scene parameters. Combining these approaches, we propose a method for scene relighting under novel views by learning a neural precomputed radiance transfer function, which implicitly handles global illumination effects using novel environment maps. Our method can be solely supervised on a set of real images of the scene under a single unknown lighting condition. To disambiguate the task during training, we tightly integrate a differentiable path tracer in the training process and propose a combination of a synthesized OLAT and a real image loss. Results show that the recovered disentanglement of scene parameters improves significantly over the current state of the art and, thus, also our re-rendering results are more realistic and accurate.

• ShAPO: Implicit Representations for Multi-Object Shape, Appearance, and Pose Optimization, ECCV2022 | [code]

  Our method studies the complex task of object-centric 3D understanding from a single RGB-D observation. As it is an ill-posed problem, existing methods suffer from low performance for both 3D shape and 6D pose and size estimation in complex multi-object scenarios with occlusions. We present ShAPO, a method for joint multi-object detection, 3D textured reconstruction, 6D object pose and size estimation. Key to ShAPO is a single-shot pipeline to regress shape, appearance and pose latent codes along with the masks of each object instance, which is then further refined in a sparse-to-dense fashion. A novel disentangled shape and appearance database of priors is first learned to embed objects in their respective shape and appearance space. We also propose a novel, octree-based differentiable optimization step, allowing us to further improve object shape, pose and appearance simultaneously under the learned latent space, in an analysis-by-synthesis fashion. Our novel joint implicit textured object representation allows us to accurately identify and reconstruct novel unseen objects without having access to their 3D meshes. Through extensive experiments, we show that our method, trained on simulated indoor scenes, accurately regresses the shape, appearance and pose of novel objects in the real-world with minimal fine-tuning. Our method significantly out-performs all baselines on the NOCS dataset with an 8% absolute improvement in mAP for 6D pose estimation.

• GAUDI: A Neural Architect for Immersive 3D Scene Generation | [code]

  We introduce GAUDI, a generative model capable of capturing the distribution of complex and realistic 3D scenes that can be rendered immersively from a moving camera. We tackle this challenging problem with a scalable yet powerful approach, where we first optimize a latent representation that disentangles radiance fields and camera poses. This latent representation is then used to learn a generative model that enables both unconditional and conditional generation of 3D scenes. Our model generalizes previous works that focus on single objects by removing the assumption that the camera pose distribution can be shared across samples. We show that GAUDI obtains state-of-the-art performance in the unconditional generative setting across multiple datasets and allows for conditional generation of 3D scenes given conditioning variables like sparse image observations or text that describes the scene.

• AlignSDF: Pose-Aligned Signed Distance Fields for Hand-Object Reconstruction, ECCV2022 | [code]

  Recent work achieved impressive progress towards joint reconstruction of hands and manipulated objects from monocular color images. Existing methods focus on two alternative representations in terms of either parametric meshes or signed distance fields (SDFs). On one side, parametric models can benefit from prior knowledge at the cost of limited shape deformations and mesh resolutions. Mesh models, hence, may fail to precisely reconstruct details such as contact surfaces of hands and objects. SDF-based methods, on the other side, can represent arbitrary details but are lacking explicit priors. In this work we aim to improve SDF models using priors provided by parametric representations. In particular, we propose a joint learning framework that disentangles the pose and the shape. We obtain hand and object poses from parametric models and use them to align SDFs in 3D space. We show that such aligned SDFs better focus on reconstructing shape details and improve reconstruction accuracy both for hands and objects. We evaluate our method and demonstrate significant improvements over the state of the art on the challenging ObMan and DexYCB benchmarks.

• Neural Green’s function for Laplacian systems, Computer & Graphics | [code]

  Solving linear system of equations stemming from Laplacian operators is at the heart of a wide range of applications. Due to the sparsity of the linear systems, iterative solvers such as Conjugate Gradient and Multigrid are usually employed when the solution has a large number of degrees of freedom. These iterative solvers can be seen as sparse approximations of the Green’s function for the Laplacian operator. In this paper we propose a machine learning approach that regresses a Green’s function from boundary conditions. This is enabled by a Green’s function that can be effectively represented in a multi-scale fashion, drastically reducing the cost associated with a dense matrix representation. Additionally, since the Green’s function is solely dependent on boundary conditions, training the proposed neural network does not require sampling the right-hand side of the linear system. We show results that our method outperforms state of the art Conjugate Gradient and Multigrid methods.

• Deforming Radiance Fields with Cages, ECCV2022 | [code]

  Recent advances in radiance fields enable photorealistic rendering of static or dynamic 3D scenes, but still do not support explicit deformation that is used for scene manipulation or animation. In this paper, we propose a method that enables a new type of deformation of the radiance field: free-form radiance field deformation. We use a triangular mesh that encloses the foreground object called cage as an interface, and by manipulating the cage vertices, our approach enables the free-form deformation of the radiance field. The core of our approach is cage-based deformation which is commonly used in mesh deformation. We propose a novel formulation to extend it to the radiance field, which maps the position and the view direction of the sampling points from the deformed space to the canonical space, thus enabling the rendering of the deformed scene. The deformation results of the synthetic datasets and the real-world datasets demonstrate the effectiveness of our approach.

• NeuMesh: Learning Disentangled Neural Mesh-based Implicit Field for Geometry and Texture Editing, ECCV2022(oral) | [code]

  Very recently neural implicit rendering techniques have been rapidly evolved and shown great advantages in novel view synthesis and 3D scene reconstruction. However, existing neural rendering methods for editing purposes offer limited functionality, e.g., rigid transformation, or not applicable for fine-grained editing for general objects from daily lives. In this paper, we present a novel mesh-based representation by encoding the neural implicit field with disentangled geometry and texture codes on mesh vertices, which facilitates a set of editing functionalities, including mesh-guided geometry editing, designated texture editing with texture swapping, filling and painting operations. To this end, we develop several techniques including learnable sign indicators to magnify spatial distinguishability of mesh-based representation, distillation and fine-tuning mechanism to make a steady convergence, and the spatial-aware optimization strategy to realize precise texture editing. Extensive experiments and editing examples on both real and synthetic data demonstrate the superiority of our method on representation quality and editing ability. Code is available on the project webpage: this https URL.

• On the Learnability of Physical Concepts: Can a Neural Network Understand What's Real? | [code]

  We revisit the classic signal-to-symbol barrier in light of the remarkable ability of deep neural networks to generate realistic synthetic data. DeepFakes and spoofing highlight the feebleness of the link between physical reality and its abstract representation, whether learned by a digital computer or a biological agent. Starting from a widely applicable definition of abstract concept, we show that standard feed-forward architectures cannot capture but trivial concepts, regardless of the number of weights and the amount of training data, despite being extremely effective classifiers. On the other hand, architectures that incorporate recursion can represent a significantly larger class of concepts, but may still be unable to learn them from a finite dataset. We qualitatively describe the class of concepts that can be "understood" by modern architectures trained with variants of stochastic gradient descent, using a (free energy) Lagrangian to measure information complexity. Even if a concept has been understood, however, a network has no means of communicating its understanding to an external agent, except through continuous interaction and validation. We then characterize physical objects as abstract concepts and use the previous analysis to show that physical objects can be encoded by finite architectures. However, to understand physical concepts, sensors must provide persistently exciting observations, for which the ability to control the data acquisition process is essential (active perception). The importance of control depends on the modality, benefiting visual more than acoustic or chemical perception. Finally, we conclude that binding physical entities to digital identities is possible in finite time with finite resources, solving in principle the signal-to-symbol barrier problem, but we highlight the need for continuous validation.

Previous weeks

• Plenoxels: Radiance Fields without Neural Networks, CVPR2022(oral) | [code]

  We introduce Plenoxels (plenoptic voxels), a system for photorealistic view synthesis. Plenoxels represent a scene as a sparse 3D grid with spherical harmonics. This representation can be optimized from calibrated images via gradient methods and regularization without any neural components. On standard, benchmark tasks, Plenoxels are optimized two orders of magnitude faster than Neural Radiance Fields with no loss in visual quality.

• Urban Radiance Fields, CVPR2022 | [code]

  The goal of this work is to perform 3D reconstruction and novel view synthesis from data captured by scanning platforms commonly deployed for world mapping in urban outdoor environments (e.g., Street View). Given a sequence of posed RGB images and lidar sweeps acquired by cameras and scanners moving through an outdoor scene, we produce a model from which 3D surfaces can be extracted and novel RGB images can be synthesized. Our approach extends Neural Radiance Fields, which has been demonstrated to synthesize realistic novel images for small scenes in controlled settings, with new methods for leveraging asynchronously captured lidar data, for addressing exposure variation between captured images, and for leveraging predicted image segmentations to supervise densities on rays pointing at the sky. Each of these three extensions provides significant performance improvements in experiments on Street View data. Our system produces state-of-the-art 3D surface reconstructions and synthesizes higher quality novel views in comparison to both traditional methods (e.g.~COLMAP) and recent neural representations (e.g.~Mip-NeRF).

• NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis, ECCV2020 | [code]

  We present a method that achieves state-of-the-art results for synthesizing novel views of complex scenes by optimizing an underlying continuous volumetric scene function using a sparse set of input views. Our algorithm represents a scene using a fully-connected (non-convolutional) deep network, whose input is a single continuous 5D coordinate (spatial location (x,y,z) and viewing direction (θ,ϕ)) and whose output is the volume density and view-dependent emitted radiance at that spatial location. We synthesize views by querying 5D coordinates along camera rays and use classic volume rendering techniques to project the output colors and densities into an image. Because volume rendering is naturally differentiable, the only input required to optimize our representation is a set of images with known camera poses. We describe how to effectively optimize neural radiance fields to render photorealistic novel views of scenes with complicated geometry and appearance, and demonstrate results that outperform prior work on neural rendering and view synthesis. View synthesis results are best viewed as videos, so we urge readers to view our supplementary video for convincing comparisons.

• Neural Sparse Voxel Fields, NeurIPS2020 | [code]

  We introduce Neural Sparse Voxel Fields (NSVF), a new neural scene representation for fast and high-quality free-viewpoint rendering. NSVF defines a set of voxel-bounded implicit fields organized in a sparse voxel octree to model local properties in each cell. We progressively learn the underlying voxel structures with a diffentiable ray-marching operation from only a set of posed RGB images. With the sparse voxel octree structure, rendering novel views can be accelerated by skipping the voxels containing no relevant scene content. Our method is over 10 times faster than the state-of-the-art (namely, NeRF (Mildenhall et al., 2020)) at inference time while achieving higher quality results. Furthermore, by utilizing an explicit sparse voxel representation, our method can easily be applied to scene editing and scene composition. We also demonstrate several challenging tasks, including multi-scene learning, free-viewpoint rendering of a moving human, and large-scale scene rendering.

• AutoInt: Automatic Integration for Fast Neural Volume Rendering, CVPR2021 | [code]

  Numerical integration is a foundational technique in scientific computing and is at the core of many computer vision applications. Among these applications, implicit neural volume rendering has recently been proposed as a new paradigm for view synthesis, achieving photorealistic image quality. However, a fundamental obstacle to making these methods practical is the extreme computational and memory requirements caused by the required volume integrations along the rendered rays during training and inference. Millions of rays, each requiring hundreds of forward passes through a neural network are needed to approximate those integrations with Monte Carlo sampling. Here, we propose automatic integration, a new framework for learning efficient, closed-form solutions to integrals using implicit neural representation networks. For training, we instantiate the computational graph corresponding to the derivative of the implicit neural representation. The graph is fitted to the signal to integrate. After optimization, we reassemble the graph to obtain a network that represents the antiderivative. By the fundamental theorem of calculus, this enables the calculation of any definite integral in two evaluations of the network. Using this approach, we demonstrate a greater than 10× improvement in computation requirements, enabling fast neural volume rendering.

• DeRF: Decomposed Radiance Fields | [code]

  With the advent of Neural Radiance Fields (NeRF), neural networks can now render novel views of a 3D scene with quality that fools the human eye. Yet, generating these images is very computationally intensive, limiting their applicability in practical scenarios. In this paper, we propose a technique based on spatial decomposition capable of mitigating this issue. Our key observation is that there are diminishing returns in employing larger (deeper and/or wider) networks. Hence, we propose to spatially decompose a scene and dedicate smaller networks for each decomposed part. When working together, these networks can render the whole scene. This allows us near-constant inference time regardless of the number of decomposed parts. Moreover, we show that a Voronoi spatial decomposition is preferable for this purpose, as it is provably compatible with the Painter's Algorithm for efficient and GPU-friendly rendering. Our experiments show that for real-world scenes, our method provides up to 3x more efficient inference than NeRF (with the same rendering quality), or an improvement of up to 1.0~dB in PSNR (for the same inference cost).

• DONeRF: Towards Real-Time Rendering of Compact Neural Radiance Fields using Depth Oracle Networks, CGF2021 | [code]

  The recent research explosion around Neural Radiance Fields (NeRFs) shows that there is immense potential for implicitly storing scene and lighting information in neural networks, e.g., for novel view generation. However, one major limitation preventing the widespread use of NeRFs is the prohibitive computational cost of excessive network evaluations along each view ray, requiring dozens of petaFLOPS when aiming for real-time rendering on current devices. We show that the number of samples required for each view ray can be significantly reduced when local samples are placed around surfaces in the scene. To this end, we propose a depth oracle network, which predicts ray sample locations for each view ray with a single network evaluation. We show that using a classification network around logarithmically discretized and spherically warped depth values is essential to encode surface locations rather than directly estimating depth. The combination of these techniques leads to DONeRF, a dual network design with a depth oracle network as a first step and a locally sampled shading network for ray accumulation. With our design, we reduce the inference costs by up to 48x compared to NeRF. Using an off-the-shelf inference API in combination with simple compute kernels, we are the first to render raymarching-based neural representations at interactive frame rates (15 frames per second at 800x800) on a single GPU. At the same time, since we focus on the important parts of the scene around surfaces, we achieve equal or better quality compared to NeRF.

• FastNeRF: High-Fidelity Neural Rendering at 200FPS, ICCV2021 | [code]

  Recent work on Neural Radiance Fields (NeRF) showed how neural networks can be used to encode complex 3D environments that can be rendered photorealistically from novel viewpoints. Rendering these images is very computationally demanding and recent improvements are still a long way from enabling interactive rates, even on high-end hardware. Motivated by scenarios on mobile and mixed reality devices, we propose FastNeRF, the first NeRF-based system capable of rendering high fidelity photorealistic images at 200Hz on a high-end consumer GPU. The core of our method is a graphics-inspired factorization that allows for (i) compactly caching a deep radiance map at each position in space, (ii) efficiently querying that map using ray directions to estimate the pixel values in the rendered image. Extensive experiments show that the proposed method is 3000 times faster than the original NeRF algorithm and at least an order of magnitude faster than existing work on accelerating NeRF, while maintaining visual quality and extensibility.

• KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs , ICCV2021 | [code]

  NeRF synthesizes novel views of a scene with unprecedented quality by fitting a neural radiance field to RGB images. However, NeRF requires querying a deep Multi-Layer Perceptron (MLP) millions of times, leading to slow rendering times, even on modern GPUs. In this paper, we demonstrate that real-time rendering is possible by utilizing thousands of tiny MLPs instead of one single large MLP. In our setting, each individual MLP only needs to represent parts of the scene, thus smaller and faster-to-evaluate MLPs can be used. By combining this divide-and-conquer strategy with further optimizations, rendering is accelerated by three orders of magnitude compared to the original NeRF model without incurring high storage costs. Further, using teacher-student distillation for training, we show that this speed-up can be achieved without sacrificing visual quality.

• PlenOctrees for Real-time Rendering of Neural Radiance Fields, ICCV2021(oral) | [code]

  Real-time performance is achieved by pre-tabulating the NeRF into an octree-based radiance field that we call PlenOctrees. In order to preserve view-dependent effects such as specularities, we propose to encode appearances via closed-form spherical basis functions. Specifically, we show that it is possible to train NeRFs to predict a spherical harmonic representation of radiance, removing the viewing direction as input to the neural network. Furthermore, we show that our PlenOctrees can be directly optimized to further minimize the reconstruction loss, which leads to equal or better quality than competing methods. We further show that this octree optimization step can be used to accelerate the training time, as we no longer need to wait for the NeRF training to converge fully. Our real-time neural rendering approach may potentially enable new applications such as 6-DOF industrial and product visualizations, as well as next generation AR/VR systems.

• Mixture of Volumetric Primitives for Efficient Neural Rendering, SIGGRAPH2021 | [code]

  Real-time rendering and animation of humans is a core function in games, movies, and telepresence applications. Existing methods have a number of drawbacks we aim to address with our work. Triangle meshes have difficulty modeling thin structures like hair, volumetric representations like Neural Volumes are too low-resolution given a reasonable memory budget, and high-resolution implicit representations like Neural Radiance Fields are too slow for use in real-time applications. We present Mixture of Volumetric Primitives (MVP), a representation for rendering dynamic 3D content that combines the completeness of volumetric representations with the efficiency of primitive-based rendering, e.g., point-based or mesh-based methods. Our approach achieves this by leveraging spatially shared computation with a deconvolutional architecture and by minimizing computation in empty regions of space with volumetric primitives that can move to cover only occupied regions. Our parameterization supports the integration of correspondence and tracking constraints, while being robust to areas where classical tracking fails, such as around thin or translucent structures and areas with large topological variability. MVP is a hybrid that generalizes both volumetric and primitive-based representations. Through a series of extensive experiments we demonstrate that it inherits the strengths of each, while avoiding many of their limitations. We also compare our approach to several state-of-the-art methods and demonstrate that MVP produces superior results in terms of quality and runtime performance.

• Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering, NeurIPS2021(spotlight) | [code]

  Inferring representations of 3D scenes from 2D observations is a fundamental problem of computer graphics, computer vision, and artificial intelligence. Emerging 3D-structured neural scene representations are a promising approach to 3D scene understanding. In this work, we propose a novel neural scene representation, Light Field Networks or LFNs, which represent both geometry and appearance of the underlying 3D scene in a 360-degree, four-dimensional light field parameterized via a neural implicit representation. Rendering a ray from an LFN requires only a single network evaluation, as opposed to hundreds of evaluations per ray for ray-marching or volumetric based renderers in 3D-structured neural scene representations. In the setting of simple scenes, we leverage meta-learning to learn a prior over LFNs that enables multi-view consistent light field reconstruction from as little as a single image observation. This results in dramatic reductions in time and memory complexity, and enables real-time rendering. The cost of storing a 360-degree light field via an LFN is two orders of magnitude lower than conventional methods such as the Lumigraph. Utilizing the analytical differentiability of neural implicit representations and a novel parameterization of light space, we further demonstrate the extraction of sparse depth maps from LFNs.

• Depth-supervised NeRF: Fewer Views and Faster Training for Free, CVPR2022 | [code]

  A commonly observed failure mode of Neural Radiance Field (NeRF) is fitting incorrect geometries when given an insufficient number of input views. One potential reason is that standard volumetric rendering does not enforce the constraint that most of a scene's geometry consist of empty space and opaque surfaces. We formalize the above assumption through DS-NeRF (Depth-supervised Neural Radiance Fields), a loss for learning radiance fields that takes advantage of readily-available depth supervision. We leverage the fact that current NeRF pipelines require images with known camera poses that are typically estimated by running structure-from-motion (SFM). Crucially, SFM also produces sparse 3D points that can be used as "free" depth supervision during training: we add a loss to encourage the distribution of a ray's terminating depth matches a given 3D keypoint, incorporating depth uncertainty. DS-NeRF can render better images given fewer training views while training 2-3x faster. Further, we show that our loss is compatible with other recently proposed NeRF methods, demonstrating that depth is a cheap and easily digestible supervisory signal. And finally, we find that DS-NeRF can support other types of depth supervision such as scanned depth sensors and RGB-D reconstruction outputs.

• Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction, CVPR2022(oral) | [code]

  We present a super-fast convergence approach to reconstructing the per-scene radiance field from a set of images that capture the scene with known poses. This task, which is often applied to novel view synthesis, is recently revolutionized by Neural Radiance Field (NeRF) for its state-of-the-art quality and flexibility. However, NeRF and its variants require a lengthy training time ranging from hours to days for a single scene. In contrast, our approach achieves NeRF-comparable quality and converges rapidly from scratch in less than 15 minutes with a single GPU. We adopt a representation consisting of a density voxel grid for scene geometry and a feature voxel grid with a shallow network for complex view-dependent appearance. Modeling with explicit and discretized volume representations is not new, but we propose two simple yet non-trivial techniques that contribute to fast convergence speed and high-quality output. First, we introduce the post-activation interpolation on voxel density, which is capable of producing sharp surfaces in lower grid resolution. Second, direct voxel density optimization is prone to suboptimal geometry solutions, so we robustify the optimization process by imposing several priors. Finally, evaluation on five inward-facing benchmarks shows that our method matches, if not surpasses, NeRF's quality, yet it only takes about 15 minutes to train from scratch for a new scene.

• NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collections, CVPR2021 | [code]

  We present a learning-based method for synthesizing novel views of complex scenes using only unstructured collections of in-the-wild photographs. We build on Neural Radiance Fields (NeRF), which uses the weights of a multilayer perceptron to model the density and color of a scene as a function of 3D coordinates. While NeRF works well on images of static subjects captured under controlled settings, it is incapable of modeling many ubiquitous, real-world phenomena in uncontrolled images, such as variable illumination or transient occluders. We introduce a series of extensions to NeRF to address these issues, thereby enabling accurate reconstructions from unstructured image collections taken from the internet. We apply our system, dubbed NeRF-W, to internet photo collections of famous landmarks, and demonstrate temporally consistent novel view renderings that are significantly closer to photorealism than the prior state of the art.

• Ha-NeRF: Hallucinated Neural Radiance Fields in the Wild, CVPR2022 | [code]

  Neural Radiance Fields (NeRF) has recently gained popularity for its impressive novel view synthesis ability. This paper studies the problem of hallucinated NeRF: i.e., recovering a realistic NeRF at a different time of day from a group of tourism images. Existing solutions adopt NeRF with a controllable appearance embedding to render novel views under various conditions, but they cannot render view-consistent images with an unseen appearance. To solve this problem, we present an end-to-end framework for constructing a hallucinated NeRF, dubbed as Ha-NeRF. Specifically, we propose an appearance hallucination module to handle time-varying appearances and transfer them to novel views. Considering the complex occlusions of tourism images, we introduce an anti-occlusion module to decompose the static subjects for visibility accurately. Experimental results on synthetic data and real tourism photo collections demonstrate that our method can hallucinate the desired appearances and render occlusion-free images from different views.

• Nerfies: Deformable Neural Radiance Fields, ICCV2021 | [code]

  We present the first method capable of photorealistically reconstructing deformable scenes using photos/videos captured casually from mobile phones. Our approach augments neural radiance fields (NeRF) by optimizing an additional continuous volumetric deformation field that warps each observed point into a canonical 5D NeRF. We observe that these NeRF-like deformation fields are prone to local minima, and propose a coarse-to-fine optimization method for coordinate-based models that allows for more robust optimization. By adapting principles from geometry processing and physical simulation to NeRF-like models, we propose an elastic regularization of the deformation field that further improves robustness. We show that our method can turn casually captured selfie photos/videos into deformable NeRF models that allow for photorealistic renderings of the subject from arbitrary viewpoints, which we dub "nerfies." We evaluate our method by collecting time-synchronized data using a rig with two mobile phones, yielding train/validation images of the same pose at different viewpoints. We show that our method faithfully reconstructs non-rigidly deforming scenes and reproduces unseen views with high fidelity.

• D-NeRF: Neural Radiance Fields for Dynamic Scenes, CVPR2021 | [code]

  Neural rendering techniques combining machine learning with geometric reasoning have arisen as one of the most promising approaches for synthesizing novel views of a scene from a sparse set of images. Among these, stands out the Neural radiance fields (NeRF), which trains a deep network to map 5D input coordinates (representing spatial location and viewing direction) into a volume density and view-dependent emitted radiance. However, despite achieving an unprecedented level of photorealism on the generated images, NeRF is only applicable to static scenes, where the same spatial location can be queried from different images. In this paper we introduce D-NeRF, a method that extends neural radiance fields to a dynamic domain, allowing to reconstruct and render novel images of objects under rigid and non-rigid motions from a \emph{single} camera moving around the scene. For this purpose we consider time as an additional input to the system, and split the learning process in two main stages: one that encodes the scene into a canonical space and another that maps this canonical representation into the deformed scene at a particular time. Both mappings are simultaneously learned using fully-connected networks. Once the networks are trained, D-NeRF can render novel images, controlling both the camera view and the time variable, and thus, the object movement. We demonstrate the effectiveness of our approach on scenes with objects under rigid, articulated and non-rigid motions. Code, model weights and the dynamic scenes dataset will be released.

• Dynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstruction, CVPR2021 | [code]

  We present dynamic neural radiance fields for modeling the appearance and dynamics of a human face. Digitally modeling and reconstructing a talking human is a key building-block for a variety of applications. Especially, for telepresence applications in AR or VR, a faithful reproduction of the appearance including novel viewpoint or head-poses is required. In contrast to state-of-the-art approaches that model the geometry and material properties explicitly, or are purely image-based, we introduce an implicit representation of the head based on scene representation networks. To handle the dynamics of the face, we combine our scene representation network with a low-dimensional morphable model which provides explicit control over pose and expressions. We use volumetric rendering to generate images from this hybrid representation and demonstrate that such a dynamic neural scene representation can be learned from monocular input data only, without the need of a specialized capture setup. In our experiments, we show that this learned volumetric representation allows for photo-realistic image generation that surpasses the quality of state-of-the-art video-based reenactment methods.

• Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Deforming Scene from Monocular Video,, ICCV2021 | [code]

  We present Non-Rigid Neural Radiance Fields (NR-NeRF), a reconstruction and novel view synthesis approach for general non-rigid dynamic scenes. Our approach takes RGB images of a dynamic scene as input (e.g., from a monocular video recording), and creates a high-quality space-time geometry and appearance representation. We show that a single handheld consumer-grade camera is sufficient to synthesize sophisticated renderings of a dynamic scene from novel virtual camera views, e.g. a `bullet-time' video effect. NR-NeRF disentangles the dynamic scene into a canonical volume and its deformation. Scene deformation is implemented as ray bending, where straight rays are deformed non-rigidly. We also propose a novel rigidity network to better constrain rigid regions of the scene, leading to more stable results. The ray bending and rigidity network are trained without explicit supervision. Our formulation enables dense correspondence estimation across views and time, and compelling video editing applications such as motion exaggeration. Our code will be open sourced.

• PVA: Pixel-aligned Volumetric Avatars, CVPR2021 | [code]

  Acquisition and rendering of photorealistic human heads is a highly challenging research problem of particular importance for virtual telepresence. Currently, the highest quality is achieved by volumetric approaches trained in a person-specific manner on multi-view data. These models better represent fine structure, such as hair, compared to simpler mesh-based models. Volumetric models typically employ a global code to represent facial expressions, such that they can be driven by a small set of animation parameters. While such architectures achieve impressive rendering quality, they can not easily be extended to the multi-identity setting. In this paper, we devise a novel approach for predicting volumetric avatars of the human head given just a small number of inputs. We enable generalization across identities by a novel parameterization that combines neural radiance fields with local, pixel-aligned features extracted directly from the inputs, thus side-stepping the need for very deep or complex networks. Our approach is trained in an end-to-end manner solely based on a photometric rerendering loss without requiring explicit 3D supervision.We demonstrate that our approach outperforms the existing state of the art in terms of quality and is able to generate faithful facial expressions in a multi-identity setting.

• Neural Articulated Radiance Field, ICCV2021 | [code]

  We present Neural Articulated Radiance Field (NARF), a novel deformable 3D representation for articulated objects learned from images. While recent advances in 3D implicit representation have made it possible to learn models of complex objects, learning pose-controllable representations of articulated objects remains a challenge, as current methods require 3D shape supervision and are unable to render appearance. In formulating an implicit representation of 3D articulated objects, our method considers only the rigid transformation of the most relevant object part in solving for the radiance field at each 3D location. In this way, the proposed method represents pose-dependent changes without significantly increasing the computational complexity. NARF is fully differentiable and can be trained from images with pose annotations. Moreover, through the use of an autoencoder, it can learn appearance variations over multiple instances of an object class. Experiments show that the proposed method is efficient and can generalize well to novel poses.

• CLA-NeRF: Category-Level Articulated Neural Radiance Field, ICRA2022 | [code]

  We propose CLA-NeRF -- a Category-Level Articulated Neural Radiance Field that can perform view synthesis, part segmentation, and articulated pose estimation. CLA-NeRF is trained at the object category level using no CAD models and no depth, but a set of RGB images with ground truth camera poses and part segments. During inference, it only takes a few RGB views (i.e., few-shot) of an unseen 3D object instance within the known category to infer the object part segmentation and the neural radiance field. Given an articulated pose as input, CLA-NeRF can perform articulation-aware volume rendering to generate the corresponding RGB image at any camera pose. Moreover, the articulated pose of an object can be estimated via inverse rendering. In our experiments, we evaluate the framework across five categories on both synthetic and real-world data. In all cases, our method shows realistic deformation results and accurate articulated pose estimation. We believe that both few-shot articulated object rendering and articulated pose estimation open doors for robots to perceive and interact with unseen articulated objects.

• Animatable Neural Radiance Fields for Human Body Modeling, ICCV2021 | [code]

  This paper addresses the challenge of reconstructing an animatable human model from a multi-view video. Some recent works have proposed to decompose a non-rigidly deforming scene into a canonical neural radiance field and a set of deformation fields that map observation-space points to the canonical space, thereby enabling them to learn the dynamic scene from images. However, they represent the deformation field as translational vector field or SE(3) field, which makes the optimization highly under-constrained. Moreover, these representations cannot be explicitly controlled by input motions. Instead, we introduce neural blend weight fields to produce the deformation fields. Based on the skeleton-driven deformation, blend weight fields are used with 3D human skeletons to generate observation-to-canonical and canonical-to-observation correspondences. Since 3D human skeletons are more observable, they can regularize the learning of deformation fields. Moreover, the learned blend weight fields can be combined with input skeletal motions to generate new deformation fields to animate the human model. Experiments show that our approach significantly outperforms recent human synthesis methods. The code will be available at https://zju3dv.github.io/animatable_nerf/.

• Neural Actor: Neural Free-view Synthesis of Human Actors with Pose Control, SIGSIGGRAPH Asia 2021 | [code]

  We propose Neural Actor (NA), a new method for high-quality synthesis of humans from arbitrary viewpoints and under arbitrary controllable poses. Our method is built upon recent neural scene representation and rendering works which learn representations of geometry and appearance from only 2D images. While existing works demonstrated compelling rendering of static scenes and playback of dynamic scenes, photo-realistic reconstruction and rendering of humans with neural implicit methods, in particular under user-controlled novel poses, is still difficult. To address this problem, we utilize a coarse body model as the proxy to unwarp the surrounding 3D space into a canonical pose. A neural radiance field learns pose-dependent geometric deformations and pose- and view-dependent appearance effects in the canonical space from multi-view video input. To synthesize novel views of high fidelity dynamic geometry and appearance, we leverage 2D texture maps defined on the body model as latent variables for predicting residual deformations and the dynamic appearance. Experiments demonstrate that our method achieves better quality than the state-of-the-arts on playback as well as novel pose synthesis, and can even generalize well to new poses that starkly differ from the training poses. Furthermore, our method also supports body shape control of the synthesized results.

• Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenes, CVPR2021 | [code]

  We present a method to perform novel view and time synthesis of dynamic scenes, requiring only a monocular video with known camera poses as input. To do this, we introduce Neural Scene Flow Fields, a new representation that models the dynamic scene as a time-variant continuous function of appearance, geometry, and 3D scene motion. Our representation is optimized through a neural network to fit the observed input views. We show that our representation can be used for complex dynamic scenes, including thin structures, view-dependent effects, and natural degrees of motion. We conduct a number of experiments that demonstrate our approach significantly outperforms recent monocular view synthesis methods, and show qualitative results of space-time view synthesis on a variety of real-world videos.

• Neural Body: Implicit Neural Representations with Structured Latent Codes for Novel View Synthesis of Dynamic Humans, CVPR2021 | [code]

  This paper addresses the challenge of novel view synthesis for a human performer from a very sparse set of camera views. Some recent works have shown that learning implicit neural representations of 3D scenes achieves remarkable view synthesis quality given dense input views. However, the representation learning will be ill-posed if the views are highly sparse. To solve this ill-posed problem, our key idea is to integrate observations over video frames. To this end, we propose Neural Body, a new human body representation which assumes that the learned neural representations at different frames share the same set of latent codes anchored to a deformable mesh, so that the observations across frames can be naturally integrated. The deformable mesh also provides geometric guidance for the network to learn 3D representations more efficiently. To evaluate our approach, we create a multi-view dataset named ZJU-MoCap that captures performers with complex motions. Experiments on ZJU-MoCap show that our approach outperforms prior works by a large margin in terms of novel view synthesis quality. We also demonstrate the capability of our approach to reconstruct a moving person from a monocular video on the People-Snapshot dataset.

• Neural 3D Video Synthesis from Multi-view Video, CVPR2022(oral) | [code]

  We propose a novel approach for 3D video synthesis that is able to represent multi-view video recordings of a dynamic real-world scene in a compact, yet expressive representation that enables high-quality view synthesis and motion interpolation. Our approach takes the high quality and compactness of static neural radiance fields in a new direction: to a model-free, dynamic setting. At the core of our approach is a novel time-conditioned neural radiance fields that represents scene dynamics using a set of compact latent codes. To exploit the fact that changes between adjacent frames of a video are typically small and locally consistent, we propose two novel strategies for efficient training of our neural network: 1) An efficient hierarchical training scheme, and 2) an importance sampling strategy that selects the next rays for training based on the temporal variation of the input videos. In combination, these two strategies significantly boost the training speed, lead to fast convergence of the training process, and enable high quality results. Our learned representation is highly compact and able to represent a 10 second 30 FPS multi-view video recording by 18 cameras with a model size of just 28MB. We demonstrate that our method can render high-fidelity wide-angle novel views at over 1K resolution, even for highly complex and dynamic scenes. We perform an extensive qualitative and quantitative evaluation that shows that our approach outperforms the current state of the art. Project website: https://neural-3d-video.github.io.

• Dynamic View Synthesis from Dynamic Monocular Video, ICCV2021 | [code]

  We present an algorithm for generating novel views at arbitrary viewpoints and any input time step given a monocular video of a dynamic scene. Our work builds upon recent advances in neural implicit representation and uses continuous and differentiable functions for modeling the time-varying structure and the appearance of the scene. We jointly train a time-invariant static NeRF and a time-varying dynamic NeRF, and learn how to blend the results in an unsupervised manner. However, learning this implicit function from a single video is highly ill-posed (with infinitely many solutions that match the input video). To resolve the ambiguity, we introduce regularization losses to encourage a more physically plausible solution. We show extensive quantitative and qualitative results of dynamic view synthesis from casually captured videos.

• GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis, NeurIPS2020 | [code]

  While 2D generative adversarial networks have enabled high-resolution image synthesis, they largely lack an understanding of the 3D world and the image formation process. Thus, they do not provide precise control over camera viewpoint or object pose. To address this problem, several recent approaches leverage intermediate voxel-based representations in combination with differentiable rendering. However, existing methods either produce low image resolution or fall short in disentangling camera and scene properties, eg, the object identity may vary with the viewpoint. In this paper, we propose a generative model for radiance fields which have recently proven successful for novel view synthesis of a single scene. In contrast to voxel-based representations, radiance fields are not confined to a coarse discretization of the 3D space, yet allow for disentangling camera and scene properties while degrading gracefully in the presence of reconstruction ambiguity. By introducing a multi-scale patch-based discriminator, we demonstrate synthesis of high-resolution images while training our model from unposed 2D images alone. We systematically analyze our approach on several challenging synthetic and real-world datasets. Our experiments reveal that radiance fields are a powerful representation for generative image synthesis, leading to 3D consistent models that render with high fidelity.

• GRF: Learning a General Radiance Field for 3D Scene Representation and Rendering, ICCV2021(oral) | [code]

  We present a simple yet powerful neural network that implicitly represents and renders 3D objects and scenes only from 2D observations. The network models 3D geometries as a general radiance field, which takes a set of 2D images with camera poses and intrinsics as input, constructs an internal representation for each point of the 3D space, and then renders the corresponding appearance and geometry of that point viewed from an arbitrary position. The key to our approach is to learn local features for each pixel in 2D images and to then project these features to 3D points, thus yielding general and rich point representations. We additionally integrate an attention mechanism to aggregate pixel features from multiple 2D views, such that visual occlusions are implicitly taken into account. Extensive experiments demonstrate that our method can generate high-quality and realistic novel views for novel objects, unseen categories and challenging real-world scenes.

• pixelNeRF: Neural Radiance Fields from One or Few Images, CVPR2021 | [code]

  We propose pixelNeRF, a learning framework that predicts a continuous neural scene representation conditioned on one or few input images. The existing approach for constructing neural radiance fields involves optimizing the representation to every scene independently, requiring many calibrated views and significant compute time. We take a step towards resolving these shortcomings by introducing an architecture that conditions a NeRF on image inputs in a fully convolutional manner. This allows the network to be trained across multiple scenes to learn a scene prior, enabling it to perform novel view synthesis in a feed-forward manner from a sparse set of views (as few as one). Leveraging the volume rendering approach of NeRF, our model can be trained directly from images with no explicit 3D supervision. We conduct extensive experiments on ShapeNet benchmarks for single image novel view synthesis tasks with held-out objects as well as entire unseen categories. We further demonstrate the flexibility of pixelNeRF by demonstrating it on multi-object ShapeNet scenes and real scenes from the DTU dataset. In all cases, pixelNeRF outperforms current state-of-the-art baselines for novel view synthesis and single image 3D reconstruction. For the video and code, please visit the project website: this https URL

• Learned Initializations for Optimizing Coordinate-Based Neural Representations, CVPR2021 | [code]

  Coordinate-based neural representations have shown significant promise as an alternative to discrete, array-based representations for complex low dimensional signals. However, optimizing a coordinate-based network from randomly initialized weights for each new signal is inefficient. We propose applying standard meta-learning algorithms to learn the initial weight parameters for these fully-connected networks based on the underlying class of signals being represented (e.g., images of faces or 3D models of chairs). Despite requiring only a minor change in implementation, using these learned initial weights enables faster convergence during optimization and can serve as a strong prior over the signal class being modeled, resulting in better generalization when only partial observations of a given signal are available. We explore these benefits across a variety of tasks, including representing 2D images, reconstructing CT scans, and recovering 3D shapes and scenes from 2D image observations.

• pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis, CVPR2021(oral) | [code]

  We have witnessed rapid progress on 3D-aware image synthesis, leveraging recent advances in generative visual models and neural rendering. Existing approaches however fall short in two ways: first, they may lack an underlying 3D representation or rely on view-inconsistent rendering, hence synthesizing images that are not multi-view consistent; second, they often depend upon representation network architectures that are not expressive enough, and their results thus lack in image quality. We propose a novel generative model, named Periodic Implicit Generative Adversarial Networks (π-GAN or pi-GAN), for high-quality 3D-aware image synthesis. π-GAN leverages neural representations with periodic activation functions and volumetric rendering to represent scenes as view-consistent 3D representations with fine detail. The proposed approach obtains state-of-the-art results for 3D-aware image synthesis with multiple real and synthetic datasets.

• Portrait Neural Radiance Fields from a Single Image | [code]

  We present a method for estimating Neural Radiance Fields (NeRF) from a single headshot portrait. While NeRF has demonstrated high-quality view synthesis, it requires multiple images of static scenes and thus impractical for casual captures and moving subjects. In this work, we propose to pretrain the weights of a multilayer perceptron (MLP), which implicitly models the volumetric density and colors, with a meta-learning framework using a light stage portrait dataset. To improve the generalization to unseen faces, we train the MLP in the canonical coordinate space approximated by 3D face morphable models. We quantitatively evaluate the method using controlled captures and demonstrate the generalization to real portrait images, showing favorable results against state-of-the-arts.

• ShaRF: Shape-conditioned Radiance Fields from a Single View, ICML2021 | [code]

  We present a method for estimating neural scenes representations of objects given only a single image. The core of our method is the estimation of a geometric scaffold for the object and its use as a guide for the reconstruction of the underlying radiance field. Our formulation is based on a generative process that first maps a latent code to a voxelized shape, and then renders it to an image, with the object appearance being controlled by a second latent code. During inference, we optimize both the latent codes and the networks to fit a test image of a new object. The explicit disentanglement of shape and appearance allows our model to be fine-tuned given a single image. We can then render new views in a geometrically consistent manner and they represent faithfully the input object. Additionally, our method is able to generalize to images outside of the training domain (more realistic renderings and even real photographs). Finally, the inferred geometric scaffold is itself an accurate estimate of the object's 3D shape. We demonstrate in several experiments the effectiveness of our approach in both synthetic and real images.

• IBRNet: Learning Multi-View Image-Based Rendering, CVPR2021 | [code]

  We present a method that synthesizes novel views of complex scenes by interpolating a sparse set of nearby views. The core of our method is a network architecture that includes a multilayer perceptron and a ray transformer that estimates radiance and volume density at continuous 5D locations (3D spatial locations and 2D viewing directions), drawing appearance information on the fly from multiple source views. By drawing on source views at render time, our method hearkens back to classic work on image-based rendering (IBR), and allows us to render high-resolution imagery. Unlike neural scene representation work that optimizes per-scene functions for rendering, we learn a generic view interpolation function that generalizes to novel scenes. We render images using classic volume rendering, which is fully differentiable and allows us to train using only multi-view posed images as supervision. Experiments show that our method outperforms recent novel view synthesis methods that also seek to generalize to novel scenes. Further, if fine-tuned on each scene, our method is competitive with state-of-the-art single-scene neural rendering methods. Project page: this https URL

• CAMPARI: Camera-Aware Decomposed Generative Neural Radiance Fields | [code]

  Tremendous progress in deep generative models has led to photorealistic image synthesis. While achieving compelling results, most approaches operate in the two-dimensional image domain, ignoring the three-dimensional nature of our world. Several recent works therefore propose generative models which are 3D-aware, i.e., scenes are modeled in 3D and then rendered differentiably to the image plane. This leads to impressive 3D consistency, but incorporating such a bias comes at a price: the camera needs to be modeled as well. Current approaches assume fixed intrinsics and a predefined prior over camera pose ranges. As a result, parameter tuning is typically required for real-world data, and results degrade if the data distribution is not matched. Our key hypothesis is that learning a camera generator jointly with the image generator leads to a more principled approach to 3D-aware image synthesis. Further, we propose to decompose the scene into a background and foreground model, leading to more efficient and disentangled scene representations. While training from raw, unposed image collections, we learn a 3D- and camera-aware generative model which faithfully recovers not only the image but also the camera data distribution. At test time, our model generates images with explicit control over the camera as well as the shape and appearance of the scene.

• NeRF-VAE: A Geometry Aware 3D Scene Generative Model | [code]

  We propose NeRF-VAE, a 3D scene generative model that incorporates geometric structure via NeRF and differentiable volume rendering. In contrast to NeRF, our model takes into account shared structure across scenes, and is able to infer the structure of a novel scene -- without the need to re-train -- using amortized inference. NeRF-VAE's explicit 3D rendering process further contrasts previous generative models with convolution-based rendering which lacks geometric structure. Our model is a VAE that learns a distribution over radiance fields by conditioning them on a latent scene representation. We show that, once trained, NeRF-VAE is able to infer and render geometrically-consistent scenes from previously unseen 3D environments using very few input images. We further demonstrate that NeRF-VAE generalizes well to out-of-distribution cameras, while convolutional models do not. Finally, we introduce and study an attention-based conditioning mechanism of NeRF-VAE's decoder, which improves model performance.

• Unconstrained Scene Generation with Locally Conditioned Radiance Fields, ICCV2021 | [code]

  We follow an adversarial learning framework, where the generator models scenes via their radiance field, and the discriminator attempts to distinguish between images rendered from those radiance fields and images of real scenes. Conceptually, our model decomposes the radiance field of a scene into many small local radiance fields that result from conditioning on a 2D grid of latent codes W. W can be interpreted as a latent floorplan representing the scene.

• MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo, ICCV2021 | [code]

  We present MVSNeRF, a novel neural rendering approach that can efficiently reconstruct neural radiance fields for view synthesis. Unlike prior works on neural radiance fields that consider per-scene optimization on densely captured images, we propose a generic deep neural network that can reconstruct radiance fields from only three nearby input views via fast network inference. Our approach leverages plane-swept cost volumes (widely used in multi-view stereo) for geometry-aware scene reasoning, and combines this with physically based volume rendering for neural radiance field reconstruction. We train our network on real objects in the DTU dataset, and test it on three different datasets to evaluate its effectiveness and generalizability. Our approach can generalize across scenes (even indoor scenes, completely different from our training scenes of objects) and generate realistic view synthesis results using only three input images, significantly outperforming concurrent works on generalizable radiance field reconstruction. Moreover, if dense images are captured, our estimated radiance field representation can be easily fine-tuned; this leads to fast per-scene reconstruction with higher rendering quality and substantially less optimization time than NeRF.

• Stereo Radiance Fields (SRF): Learning View Synthesis from Sparse Views of Novel Scenes, CVPR2021 | [code]

  Recent neural view synthesis methods have achieved impressive quality and realism, surpassing classical pipelines which rely on multi-view reconstruction. State-of-the-Art methods, such as NeRF, are designed to learn a single scene with a neural network and require dense multi-view inputs. Testing on a new scene requires re-training from scratch, which takes 2-3 days. In this work, we introduce Stereo Radiance Fields (SRF), a neural view synthesis approach that is trained end-to-end, generalizes to new scenes, and requires only sparse views at test time. The core idea is a neural architecture inspired by classical multi-view stereo methods, which estimates surface points by finding similar image regions in stereo images. In SRF, we predict color and density for each 3D point given an encoding of its stereo correspondence in the input images. The encoding is implicitly learned by an ensemble of pair-wise similarities -- emulating classical stereo. Experiments show that SRF learns structure instead of overfitting on a scene. We train on multiple scenes of the DTU dataset and generalize to new ones without re-training, requiring only 10 sparse and spread-out views as input. We show that 10-15 minutes of fine-tuning further improve the results, achieving significantly sharper, more detailed results than scene-specific models. The code, model, and videos are available at this https URL.

• Neural Rays for Occlusion-aware Image-based Rendering, CVPR2022 | [code]

  We present a new neural representation, called Neural Ray (NeuRay), for the novel view synthesis task. Recent works construct radiance fields from image features of input views to render novel view images, which enables the generalization to new scenes. However, due to occlusions, a 3D point may be invisible to some input views. On such a 3D point, these generalization methods will include inconsistent image features from invisible views, which interfere with the radiance field construction. To solve this problem, we predict the visibility of 3D points to input views within our NeuRay representation. This visibility enables the radiance field construction to focus on visible image features, which significantly improves its rendering quality. Meanwhile, a novel consistency loss is proposed to refine the visibility in NeuRay when finetuning on a specific scene. Experiments demonstrate that our approach achieves state-of-the-art performance on the novel view synthesis task when generalizing to unseen scenes and outperforms per-scene optimization methods after finetuning.

• Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis, ICCV2021 | [code]

  We present DietNeRF, a 3D neural scene representation estimated from a few images. Neural Radiance Fields (NeRF) learn a continuous volumetric representation of a scene through multi-view consistency, and can be rendered from novel viewpoints by ray casting. While NeRF has an impressive ability to reconstruct geometry and fine details given many images, up to 100 for challenging 360° scenes, it often finds a degenerate solution to its image reconstruction objective when only a few input views are available. To improve few-shot quality, we propose DietNeRF. We introduce an auxiliary semantic consistency loss that encourages realistic renderings at novel poses. DietNeRF is trained on individual scenes to (1) correctly render given input views from the same pose, and (2) match high-level semantic attributes across different, random poses. Our semantic loss allows us to supervise DietNeRF from arbitrary poses. We extract these semantics using a pre-trained visual encoder such as CLIP, a Vision Transformer trained on hundreds of millions of diverse single-view, 2D photographs mined from the web with natural language supervision. In experiments, DietNeRF improves the perceptual quality of few-shot view synthesis when learned from scratch, can render novel views with as few as one observed image when pre-trained on a multi-view dataset, and produces plausible completions of completely unobserved regions.

• Towards Continuous Depth MPI with NeRF for Novel View Synthesis, ICCV2021 | [code]

  In this paper, we propose MINE to perform novel view synthesis and depth estimation via dense 3D reconstruction from a single image. Our approach is a continuous depth generalization of the Multiplane Images (MPI) by introducing the NEural radiance fields (NeRF). Given a single image as input, MINE predicts a 4-channel image (RGB and volume density) at arbitrary depth values to jointly reconstruct the camera frustum and fill in occluded contents. The reconstructed and inpainted frustum can then be easily rendered into novel RGB or depth views using differentiable rendering. Extensive experiments on RealEstate10K, KITTI and Flowers Light Fields show that our MINE outperforms state-of-the-art by a large margin in novel view synthesis. We also achieve competitive results in depth estimation on iBims-1 and NYU-v2 without annotated depth supervision. Our source code is available at this https URL

• TöRF: Time-of-Flight Radiance Fields for Dynamic Scene View Synthesis, NeurIPS2021 | [code]

  Neural networks can represent and accurately reconstruct radiance fields for static 3D scenes (e.g., NeRF). Several works extend these to dynamic scenes captured with monocular video, with promising performance. However, the monocular setting is known to be an under-constrained problem, and so methods rely on data-driven priors for reconstructing dynamic content. We replace these priors with measurements from a time-of-flight (ToF) camera, and introduce a neural representation based on an image formation model for continuous-wave ToF cameras. Instead of working with processed depth maps, we model the raw ToF sensor measurements to improve reconstruction quality and avoid issues with low reflectance regions, multi-path interference, and a sensor's limited unambiguous depth range. We show that this approach improves robustness of dynamic scene reconstruction to erroneous calibration and large motions, and discuss the benefits and limitations of integrating RGB+ToF sensors that are now available on modern smartphones.

• CodeNeRF: Disentangled Neural Radiance Fields for Object Categories, ICCV2021(oral) | [code]

  CodeNeRF is an implicit 3D neural representation that learns the variation of object shapes and textures across a category and can be trained, from a set of posed images, to synthesize novel views of unseen objects. Unlike the original NeRF, which is scene specific, CodeNeRF learns to disentangle shape and texture by learning separate embeddings. At test time, given a single unposed image of an unseen object, CodeNeRF jointly estimates camera viewpoint, and shape and appearance codes via optimization. Unseen objects can be reconstructed from a single image, and then rendered from new viewpoints or their shape and texture edited by varying the latent codes. We conduct experiments on the SRN benchmark, which show that CodeNeRF generalises well to unseen objects and achieves on-par performance with methods that require known camera pose at test time. Our results on real-world images demonstrate that CodeNeRF can bridge the sim-to-real gap.

• StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesis, ICLR2022 | [code]

  We propose StyleNeRF, a 3D-aware generative model for photo-realistic high-resolution image synthesis with high multi-view consistency, which can be trained on unstructured 2D images. Existing approaches either cannot synthesize high-resolution images with fine details or yield noticeable 3D-inconsistent artifacts. In addition, many of them lack control over style attributes and explicit 3D camera poses. StyleNeRF integrates the neural radiance field (NeRF) into a style-based generator to tackle the aforementioned challenges, i.e., improving rendering efficiency and 3D consistency for high-resolution image generation. We perform volume rendering only to produce a low-resolution feature map and progressively apply upsampling in 2D to address the first issue. To mitigate the inconsistencies caused by 2D upsampling, we propose multiple designs, including a better upsampler and a new regularization loss. With these designs, StyleNeRF can synthesize high-resolution images at interactive rates while preserving 3D consistency at high quality. StyleNeRF also enables control of camera poses and different levels of styles, which can generalize to unseen views. It also supports challenging tasks, including zoom-in and-out, style mixing, inversion, and semantic editing.

• NeRF in the Dark: High Dynamic Range View Synthesis from Noisy Raw Images, CVPR2022(oral) | [code]

  Neural Radiance Fields (NeRF) is a technique for high quality novel view synthesis from a collection of posed input images. Like most view synthesis methods, NeRF uses tonemapped low dynamic range (LDR) as input; these images have been processed by a lossy camera pipeline that smooths detail, clips highlights, and distorts the simple noise distribution of raw sensor data. We modify NeRF to instead train directly on linear raw images, preserving the scene's full dynamic range. By rendering raw output images from the resulting NeRF, we can perform novel high dynamic range (HDR) view synthesis tasks. In addition to changing the camera viewpoint, we can manipulate focus, exposure, and tonemapping after the fact. Although a single raw image appears significantly more noisy than a postprocessed one, we show that NeRF is highly robust to the zero-mean distribution of raw noise. When optimized over many noisy raw inputs (25-200), NeRF produces a scene representation so accurate that its rendered novel views outperform dedicated single and multi-image deep raw denoisers run on the same wide baseline input images. As a result, our method, which we call RawNeRF, can reconstruct scenes from extremely noisy images captured in near-darkness.

• iNeRF: Inverting Neural Radiance Fields for Pose Estimation, IROS2021 | [code]

  We present iNeRF, a framework that performs pose estimation by “inverting” a trained Neural Radiance Field(NeRF). NeRFs have been shown to be remarkably effective for the task of view synthesis — synthesizing photorealisticnovel views of real-world scenes or objects. In this work, we investigate whether we can apply analysis-by-synthesis with NeRF for 6DoF pose estimation – given an image, find the translation and rotation of a camera relative to a 3Dmodel. Starting from an initial pose estimate, we use gradient descent to minimize the residual between pixels rendered from an already-trained NeRF and pixels in an observed image. In our experiments, we first study 1) how to sample rays during pose refinement for iNeRF to collect informative gradients and 2) how different batch sizes ofrays affect iNeRF on a synthetic dataset. We then show that for complex real-world scenes from the LLFF dataset, iNeRF can improve NeRF by estimating the camera poses of novel images and using these images as additional trainingdata for NeRF. Finally, we show iNeRF can be combinedwith feature-based pose initialization. The approach outperforms all other RGB-based methods relying on syntheticdata on LineMOD.

• A-NeRF: Surface-free Human 3D Pose Refinement via Neural Rendering, NeurIPS2021 | [code]

  While deep learning reshaped the classical motion capture pipeline with feed-forward networks, generative models are required to recover fine alignment via iterative refinement. Unfortunately, the existing models are usually hand-crafted or learned in controlled conditions, only applicable to limited domains. We propose a method to learn a generative neural body model from unlabelled monocular videos by extending Neural Radiance Fields (NeRFs). We equip them with a skeleton to apply to time-varying and articulated motion. A key insight is that implicit models require the inverse of the forward kinematics used in explicit surface models. Our reparameterization defines spatial latent variables relative to the pose of body parts and thereby overcomes ill-posed inverse operations with an overparameterization. This enables learning volumetric body shape and appearance from scratch while jointly refining the articulated pose; all without ground truth labels for appearance, pose, or 3D shape on the input videos. When used for novel-view-synthesis and motion capture, our neural model improves accuracy on diverse datasets. Project website: this https URL .

• NeRF--: Neural Radiance Fields Without Known Camera Parameters | [code]

  Considering the problem of novel view synthesis (NVS) from only a set of 2D images, we simplify the training process of Neural Radiance Field (NeRF) on forward-facing scenes by removing the requirement of known or pre-computed camera parameters, including both intrinsics and 6DoF poses. To this end, we propose NeRF−−, with three contributions: First, we show that the camera parameters can be jointly optimised as learnable parameters with NeRF training, through a photometric reconstruction; Second, to benchmark the camera parameter estimation and the quality of novel view renderings, we introduce a new dataset of path-traced synthetic scenes, termed as Blender Forward-Facing Dataset (BLEFF); Third, we conduct extensive analyses to understand the training behaviours under various camera motions, and show that in most scenarios, the joint optimisation pipeline can recover accurate camera parameters and achieve comparable novel view synthesis quality as those trained with COLMAP pre-computed camera parameters.

• Implicit Mapping and Positioning in Real-Time, ICCV2021 | [code]

  We show for the first time that a multilayer perceptron (MLP) can serve as the only scene representation in a real-time SLAM system for a handheld RGB-D camera. Our network is trained in live operation without prior data, building a dense, scene-specific implicit 3D model of occupancy and colour which is also immediately used for tracking.

• NICE-SLAM  Neural Implicit Scalable Encoding for SLAM, CVPR2022 | [code]

  Neural implicit representations have recently shown encouraging results in various domains, including promising progress in simultaneous localization and mapping (SLAM). Nevertheless, existing methods produce over-smoothed scene reconstructions and have difficulty scaling up to large scenes. These limitations are mainly due to their simple fully-connected network architecture that does not incorporate local information in the observations. In this paper, we present NICE-SLAM, a dense SLAM system that incorporates multi-level local information by introducing a hierarchical scene representation. Optimizing this representation with pre-trained geometric priors enables detailed reconstruction on large indoor scenes. Compared to recent neural implicit SLAM systems, our approach is more scalable, efficient, and robust. Experiments on five challenging datasets demonstrate competitive results of NICE-SLAM in both mapping and tracking quality.

• GNeRF: GAN-based Neural Radiance Field without Posed Camera, ICCV2021(oral) | [code]

  We introduce GNeRF, a framework to marry Generative Adversarial Networks (GAN) with Neural Radiance Field (NeRF) reconstruction for the complex scenarios with unknown and even randomly initialized camera poses. Recent NeRF-based advances have gained popularity for remarkable realistic novel view synthesis. However, most of them heavily rely on accurate camera poses estimation, while few recent methods can only optimize the unknown camera poses in roughly forward-facing scenes with relatively short camera trajectories and require rough camera poses initialization. Differently, our GNeRF only utilizes randomly initialized poses for complex outside-in scenarios. We propose a novel two-phases end-to-end framework. The first phase takes the use of GANs into the new realm for optimizing coarse camera poses and radiance fields jointly, while the second phase refines them with additional photometric loss. We overcome local minima using a hybrid and iterative optimization scheme. Extensive experiments on a variety of synthetic and natural scenes demonstrate the effectiveness of GNeRF. More impressively, our approach outperforms the baselines favorably in those scenes with repeated patterns or even low textures that are regarded as extremely challenging before.

• BARF: Bundle-Adjusting Neural Radiance Fields, ICCV2021(oral) | [code]

  Neural Radiance Fields (NeRF) have recently gained a surge of interest within the computer vision community for its power to synthesize photorealistic novel views of real-world scenes. One limitation of NeRF, however, is its requirement of accurate camera poses to learn the scene representations. In this paper, we propose Bundle-Adjusting Neural Radiance Fields (BARF) for training NeRF from imperfect (or even unknown) camera poses — the joint problem of learning neural 3D representations and registering camera frames. We establish a theoretical connection to classical image alignment and show that coarse-to-fine registration is also applicable to NeRF. Furthermore, we show that naively applying positional encoding in NeRF has a negative impact on registration with a synthesis-based objective. Experiments on synthetic and real-world data show that BARF can effectively optimize the neural scene representations and resolve large camera pose misalignment at the same time. This enables view synthesis and localization of video sequences from unknown camera poses, opening up new avenues for visual localization systems (e.g. SLAM) and potential applications for dense 3D mapping and reconstruction.

• Self-Calibrating Neural Radiance Fields, ICCV2021 | [code]

  In this work, we propose a camera self-calibration algorithm for generic cameras with arbitrary non-linear distortions. We jointly learn the geometry of the scene and the accurate camera parameters without any calibration objects. Our camera model consists a pinhole model, radial distortion, and a generic noise model that can learn arbitrary non-linear camera distortions. While traditional self-calibration algorithms mostly rely on geometric constraints, we additionally incorporate photometric consistency. This requires learning the geometry of the scene and we use Neural Radiance Fields (NeRF). We also propose a new geometric loss function, viz., projected ray distance loss, to incorporate geometric consistency for complex non-linear camera models. We validate our approach on standard real image datasets and demonstrate our model can learn the camera intrinsics and extrinsics (pose) from scratch without COLMAP initialization. Also, we show that learning accurate camera models in differentiable manner allows us to improves PSNR over NeRF. We experimentally demonstrate that our proposed method is applicable to variants of NeRF. In addition, we use a set of images captured with a fish-eye lens to demonstrate that learning camera model jointly improves the performance significantly over the COLMAP initialization.

• NeRD: Neural Reflectance Decomposition from Image Collections, ICCV2021 | [code]

  Decomposing a scene into its shape, reflectance, and illumination is a challenging but important problem in computer vision and graphics. This problem is inherently more challenging when the illumination is not a single light source under laboratory conditions but is instead an unconstrained environmental illumination. Though recent work has shown that implicit representations can be used to model the radiance field of an object, most of these techniques only enable view synthesis and not relighting. Additionally, evaluating these radiance fields is resource and time-intensive. We propose a neural reflectance decomposition (NeRD) technique that uses physically-based rendering to decompose the scene into spatially varying BRDF material properties. In contrast to existing techniques, our input images can be captured under different illumination conditions. In addition, we also propose techniques to convert the learned reflectance volume into a relightable textured mesh enabling fast real-time rendering with novel illuminations. We demonstrate the potential of the proposed approach with experiments on both synthetic and real datasets, where we are able to obtain high-quality relightable 3D assets from image collections.

• NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesis, CVPR2021 | [code]

  We present a method that takes as input a set of images of a scene illuminated by unconstrained known lighting, and produces as output a 3D representation that can be rendered from novel viewpoints under arbitrary lighting conditions. Our method represents the scene as a continuous volumetric function parameterized as MLPs whose inputs are a 3D location and whose outputs are the following scene properties at that input location: volume density, surface normal, material parameters, distance to the first surface intersection in any direction, and visibility of the external environment in any direction. Together, these allow us to render novel views of the object under arbitrary lighting, including indirect illumination effects. The predicted visibility and surface intersection fields are critical to our model's ability to simulate direct and indirect illumination during training, because the brute-force techniques used by prior work are intractable for lighting conditions outside of controlled setups with a single light. Our method outperforms alternative approaches for recovering relightable 3D scene representations, and performs well in complex lighting settings that have posed a significant challenge to prior work.

• NeX: Real-time View Synthesis with Neural Basis Expansion, CVPR2021(oral) | [code]

  We present NeX, a new approach to novel view synthesis based on enhancements of multiplane image (MPI) that can reproduce NeXt-level view-dependent effects---in real time. Unlike traditional MPI that uses a set of simple RGBα planes, our technique models view-dependent effects by instead parameterizing each pixel as a linear combination of basis functions learned from a neural network. Moreover, we propose a hybrid implicit-explicit modeling strategy that improves upon fine detail and produces state-of-the-art results. Our method is evaluated on benchmark forward-facing datasets as well as our newly-introduced dataset designed to test the limit of view-dependent modeling with significantly more challenging effects such as the rainbow reflections on a CD. Our method achieves the best overall scores across all major metrics on these datasets with more than 1000× faster rendering time than the state of the art.

• NeRFactor: Neural Factorization of Shape and Reflectance Under an Unknown Illumination, TOG 2021 (Proc. SIGGRAPH Asia) | [code]

  We address the problem of recovering the shape and spatially-varying reflectance of an object from multi-view images (and their camera poses) of an object illuminated by one unknown lighting condition. This enables the rendering of novel views of the object under arbitrary environment lighting and editing of the object's material properties. The key to our approach, which we call Neural Radiance Factorization (NeRFactor), is to distill the volumetric geometry of a Neural Radiance Field (NeRF) [Mildenhall et al. 2020] representation of the object into a surface representation and then jointly refine the geometry while solving for the spatially-varying reflectance and environment lighting. Specifically, NeRFactor recovers 3D neural fields of surface normals, light visibility, albedo, and Bidirectional Reflectance Distribution Functions (BRDFs) without any supervision, using only a re-rendering loss, simple smoothness priors, and a data-driven BRDF prior learned from real-world BRDF measurements. By explicitly modeling light visibility, NeRFactor is able to separate shadows from albedo and synthesize realistic soft or hard shadows under arbitrary lighting conditions. NeRFactor is able to recover convincing 3D models for free-viewpoint relighting in this challenging and underconstrained capture setup for both synthetic and real scenes. Qualitative and quantitative experiments show that NeRFactor outperforms classic and deep learning-based state of the art across various tasks. Our videos, code, and data are available at people.csail.mit.edu/xiuming/projects/nerfactor/.

• NeRF++: Analyzing and Improving Neural Radiance Fields | [code]

  Neural Radiance Fields (NeRF) achieve impressive view synthesis results for a variety of capture settings, including 360 capture of bounded scenes and forward-facing capture of bounded and unbounded scenes. NeRF fits multi-layer perceptrons (MLPs) representing view-invariant opacity and view-dependent color volumes to a set of training images, and samples novel views based on volume rendering techniques. In this technical report, we first remark on radiance fields and their potential ambiguities, namely the shape-radiance ambiguity, and analyze NeRF's success in avoiding such ambiguities. Second, we address a parametrization issue involved in applying NeRF to 360 captures of objects within large-scale, unbounded 3D scenes. Our method improves view synthesis fidelity in this challenging scenario. Code is available at this https URL.

• GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fields, CVPR2021(oral) | [code]

  Deep generative models allow for photorealistic image synthesis at high resolutions. But for many applications, this is not enough: content creation also needs to be controllable. While several recent works investigate how to disentangle underlying factors of variation in the data, most of them operate in 2D and hence ignore that our world is three-dimensional. Further, only few works consider the compositional nature of scenes. Our key hypothesis is that incorporating a compositional 3D scene representation into the generative model leads to more controllable image synthesis. Representing scenes as compositional generative neural feature fields allows us to disentangle one or multiple objects from the background as well as individual objects' shapes and appearances while learning from unstructured and unposed image collections without any additional supervision. Combining this scene representation with a neural rendering pipeline yields a fast and realistic image synthesis model. As evidenced by our experiments, our model is able to disentangle individual objects and allows for translating and rotating them in the scene as well as changing the camera pose.

• Object-Centric Neural Scene Rendering | [code]

  We present a method for composing photorealistic scenes from captured images of objects. Our work builds upon neural radiance fields (NeRFs), which implicitly model the volumetric density and directionally-emitted radiance of a scene. While NeRFs synthesize realistic pictures, they only model static scenes and are closely tied to specific imaging conditions. This property makes NeRFs hard to generalize to new scenarios, including new lighting or new arrangements of objects. Instead of learning a scene radiance field as a NeRF does, we propose to learn object-centric neural scattering functions (OSFs), a representation that models per-object light transport implicitly using a lighting- and view-dependent neural network. This enables rendering scenes even when objects or lights move, without retraining. Combined with a volumetric path tracing procedure, our framework is capable of rendering both intra- and inter-object light transport effects including occlusions, specularities, shadows, and indirect illumination. We evaluate our approach on scene composition and show that it generalizes to novel illumination conditions, producing photorealistic, physically accurate renderings of multi-object scenes.

• Learning Compositional Radiance Fields of Dynamic Human Heads, CVPR2021(oral) | [code]

  Photorealistic rendering of dynamic humans is an important ability for telepresence systems, virtual shopping, synthetic data generation, and more. Recently, neural rendering methods, which combine techniques from computer graphics and machine learning, have created high-fidelity models of humans and objects. Some of these methods do not produce results with high-enough fidelity for driveable human models (Neural Volumes) whereas others have extremely long rendering times (NeRF). We propose a novel compositional 3D representation that combines the best of previous methods to produce both higher-resolution and faster results. Our representation bridges the gap between discrete and continuous volumetric representations by combining a coarse 3D-structure-aware grid of animation codes with a continuous learned scene function that maps every position and its corresponding local animation code to its view-dependent emitted radiance and local volume density. Differentiable volume rendering is employed to compute photo-realistic novel views of the human head and upper body as well as to train our novel representation end-to-end using only 2D supervision. In addition, we show that the learned dynamic radiance field can be used to synthesize novel unseen expressions based on a global animation code. Our approach achieves state-of-the-art results for synthesizing novel views of dynamic human heads and the upper body.

• Neural Scene Graphs for Dynamic Scenes, CVPR2021(oral) | [code]

  Recent implicit neural rendering methods have demonstrated that it is possible to learn accurate view synthesis for complex scenes by predicting their volumetric density and color supervised solely by a set of RGB images. However, existing methods are restricted to learning efficient representations of static scenes that encode all scene objects into a single neural network, and lack the ability to represent dynamic scenes and decompositions into individual scene objects. In this work, we present the first neural rendering method that decomposes dynamic scenes into scene graphs. We propose a learned scene graph representation, which encodes object transformation and radiance, to efficiently render novel arrangements and views of the scene. To this end, we learn implicitly encoded scenes, combined with a jointly learned latent representation to describe objects with a single implicit function. We assess the proposed method on synthetic and real automotive data, validating that our approach learns dynamic scenes -- only by observing a video of this scene -- and allows for rendering novel photo-realistic views of novel scene compositions with unseen sets of objects at unseen poses.

• Unsupervised Discovery of Object Radiance Fields, ICLR2022 | [code]

  We study the problem of inferring an object-centric scene representation from a single image, aiming to derive a representation that explains the image formation process, captures the scene's 3D nature, and is learned without supervision. Most existing methods on scene decomposition lack one or more of these characteristics, due to the fundamental challenge in integrating the complex 3D-to-2D image formation process into powerful inference schemes like deep networks. In this paper, we propose unsupervised discovery of Object Radiance Fields (uORF), integrating recent progresses in neural 3D scene representations and rendering with deep inference networks for unsupervised 3D scene decomposition. Trained on multi-view RGB images without annotations, uORF learns to decompose complex scenes with diverse, textured background from a single image. We show that uORF performs well on unsupervised 3D scene segmentation, novel view synthesis, and scene editing on three datasets.

• Learning Object-Compositional Neural Radiance Field for Editable Scene Rendering, ICCV2021 | [code]

  Implicit neural rendering techniques have shown promising results for novel view synthesis. However, existing methods usually encode the entire scene as a whole, which is generally not aware of the object identity and limits the ability to the high-level editing tasks such as moving or adding furniture. In this paper, we present a novel neural scene rendering system, which learns an object-compositional neural radiance field and produces realistic rendering with editing capability for a clustered and real-world scene. Specifically, we design a novel two-pathway architecture, in which the scene branch encodes the scene geometry and appearance, and the object branch encodes each standalone object conditioned on learnable object activation codes. To survive the training in heavily cluttered scenes, we propose a scene-guided training strategy to solve the 3D space ambiguity in the occluded regions and learn sharp boundaries for each object. Extensive experiments demonstrate that our system not only achieves competitive performance for static scene novel-view synthesis, but also produces realistic rendering for object-level editing.

• In-Place Scene Labelling and Understanding with Implicit Scene Representation, ICCV2021(oral) | [code]

  Semantic labelling is highly correlated with geometry and radiance reconstruction, as scene entities with similar shape and appearance are more likely to come from similar classes. Recent implicit neural reconstruction techniques are appealing as they do not require prior training data, but the same fully self-supervised approach is not possible for semantics because labels are human-defined properties.

• Editing Conditional Radiance Fields, ICCV2021 | [code]

  A neural radiance field (NeRF) is a scene model supporting high-quality view synthesis, optimized per scene. In this paper, we explore enabling user editing of a category-level NeRF - also known as a conditional radiance field - trained on a shape category. Specifically, we introduce a method for propagating coarse 2D user scribbles to the 3D space, to modify the color or shape of a local region. First, we propose a conditional radiance field that incorporates new modular network components, including a shape branch that is shared across object instances. Observing multiple instances of the same category, our model learns underlying part semantics without any supervision, thereby allowing the propagation of coarse 2D user scribbles to the entire 3D region (e.g., chair seat). Next, we propose a hybrid network update strategy that targets specific network components, which balances efficiency and accuracy. During user interaction, we formulate an optimization problem that both satisfies the user's constraints and preserves the original object structure. We demonstrate our approach on various editing tasks over three shape datasets and show that it outperforms prior neural editing approaches. Finally, we edit the appearance and shape of a real photograph and show that the edit propagates to extrapolated novel views.

• Editable Free-Viewpoint Video using a Layered Neural Representation, SIGGRAPH2021 | [code]

  Generating free-viewpoint videos is critical for immersive VR/AR experience but recent neural advances still lack the editing ability to manipulate the visual perception for large dynamic scenes. To fill this gap, in this paper we propose the first approach for editable photo-realistic free-viewpoint video generation for large-scale dynamic scenes using only sparse 16 cameras. The core of our approach is a new layered neural representation, where each dynamic entity including the environment itself is formulated into a space-time coherent neural layered radiance representation called ST-NeRF. Such layered representation supports fully perception and realistic manipulation of the dynamic scene whilst still supporting a free viewing experience in a wide range. In our ST-NeRF, the dynamic entity/layer is represented as continuous functions, which achieves the disentanglement of location, deformation as well as the appearance of the dynamic entity in a continuous and self-supervised manner. We propose a scene parsing 4D label map tracking to disentangle the spatial information explicitly, and a continuous deform module to disentangle the temporal motion implicitly. An object-aware volume rendering scheme is further introduced for the re-assembling of all the neural layers. We adopt a novel layered loss and motion-aware ray sampling strategy to enable efficient training for a large dynamic scene with multiple performers, Our framework further enables a variety of editing functions, i.e., manipulating the scale and location, duplicating or retiming individual neural layers to create numerous visual effects while preserving high realism. Extensive experiments demonstrate the effectiveness of our approach to achieve high-quality, photo-realistic, and editable free-viewpoint video generation for dynamic scenes.

• FiG-NeRF: Figure Ground Neural Radiance Fields for 3D Object Category Modelling, 3DV2021 | [code]

  We investigate the use of Neural Radiance Fields (NeRF) to learn high quality 3D object category models from collections of input images. In contrast to previous work, we are able to do this whilst simultaneously separating foreground objects from their varying backgrounds. We achieve this via a 2-component NeRF model, FiG-NeRF, that prefers explanation of the scene as a geometrically constant background and a deformable foreground that represents the object category. We show that this method can learn accurate 3D object category models using only photometric supervision and casually captured images of the objects. Additionally, our 2-part decomposition allows the model to perform accurate and crisp amodal segmentation. We quantitatively evaluate our method with view synthesis and image fidelity metrics, using synthetic, lab-captured, and in-the-wild data. Our results demonstrate convincing 3D object category modelling that exceed the performance of existing methods.

• NeRF-Tex: Neural Reflectance Field Textures, EGSR2021 | [code]

  We investigate the use of neural fields for modeling diverse mesoscale structures, such as fur, fabric, and grass. Instead of using classical graphics primitives to model the structure, we propose to employ a versatile volumetric primitive represented by a neural reflectance field (NeRF-Tex), which jointly models the geometry of the material and its response to lighting. The NeRF-Tex primitive can be instantiated over a base mesh to “texture” it with the desired meso and microscale appearance. We condition the reflectance field on user-defined parameters that control the appearance. A single NeRF texture thus captures an entire space of reflectance fields rather than one specific structure. This increases the gamut of appearances that can be modeled and provides a solution for combating repetitive texturing artifacts. We also demonstrate that NeRF textures naturally facilitate continuous level-of-detail rendering. Our approach unites the versatility and modeling power of neural networks with the artistic control needed for precise modeling of virtual scenes. While all our training data is currently synthetic, our work provides a recipe that can be further extended to extract complex, hard-to-model appearances from real images.

• Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields, ICCV2021(oral) | [code]

  The rendering procedure used by neural radiance fields (NeRF) samples a scene with a single ray per pixel and may therefore produce renderings that are excessively blurred or aliased when training or testing images observe scene content at different resolutions. The straightforward solution of supersampling by rendering with multiple rays per pixel is impractical for NeRF, because rendering each ray requires querying a multilayer perceptron hundreds of times. Our solution, which we call "mip-NeRF" (à la "mipmap"), extends NeRF to represent the scene at a continuously-valued scale. By efficiently rendering anti-aliased conical frustums instead of rays, mip-NeRF reduces objectionable aliasing artifacts and significantly improves NeRF's ability to represent fine details, while also being 7% faster than NeRF and half the size. Compared to NeRF, mip-NeRF reduces average error rates by 17% on the dataset presented with NeRF and by 60% on a challenging multiscale variant of that dataset that we present. mip-NeRF is also able to match the accuracy of a brute-force supersampled NeRF on our multiscale dataset while being 22x faster.

• UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction, ICCV2021(oral) | [code]

  Neural implicit 3D representations have emerged as a powerful paradigm for reconstructing surfaces from multi-view images and synthesizing novel views. Unfortunately, existing methods such as DVR or IDR require accurate per-pixel object masks as supervision. At the same time, neural radiance fields have revolutionized novel view synthesis. However, NeRF's estimated volume density does not admit accurate surface reconstruction. Our key insight is that implicit surface models and radiance fields can be formulated in a unified way, enabling both surface and volume rendering using the same model. This unified perspective enables novel, more efficient sampling procedures and the ability to reconstruct accurate surfaces without input masks. We compare our method on the DTU, BlendedMVS, and a synthetic indoor dataset. Our experiments demonstrate that we outperform NeRF in terms of reconstruction quality while performing on par with IDR without requiring masks.

• NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction, NeurIPS2021 | [code]

  We present a novel neural surface reconstruction method, called NeuS, for reconstructing objects and scenes with high fidelity from 2D image inputs. Existing neural surface reconstruction approaches, such as DVR and IDR, require foreground mask as supervision, easily get trapped in local minima, and therefore struggle with the reconstruction of objects with severe self-occlusion or thin structures. Meanwhile, recent neural methods for novel view synthesis, such as NeRF and its variants, use volume rendering to produce a neural scene representation with robustness of optimization, even for highly complex objects. However, extracting high-quality surfaces from this learned implicit representation is difficult because there are not sufficient surface constraints in the representation. In NeuS, we propose to represent a surface as the zero-level set of a signed distance function (SDF) and develop a new volume rendering method to train a neural SDF representation. We observe that the conventional volume rendering method causes inherent geometric errors (i.e. bias) for surface reconstruction, and therefore propose a new formulation that is free of bias in the first order of approximation, thus leading to more accurate surface reconstruction even without the mask supervision. Experiments on the DTU dataset and the BlendedMVS dataset show that NeuS outperforms the state-of-the-arts in high-quality surface reconstruction, especially for objects and scenes with complex structures and self-occlusion.

• Volume Rendering of Neural Implicit Surfaces, NeurIPS2021 | [code]

  Neural volume rendering became increasingly popular recently due to its success in synthesizing novel views of a scene from a sparse set of input images. So far, the geometry learned by neural volume rendering techniques was modeled using a generic density function. Furthermore, the geometry itself was extracted using an arbitrary level set of the density function leading to a noisy, often low fidelity reconstruction. The goal of this paper is to improve geometry representation and reconstruction in neural volume rendering. We achieve that by modeling the volume density as a function of the geometry. This is in contrast to previous work modeling the geometry as a function of the volume density. In more detail, we define the volume density function as Laplace's cumulative distribution function (CDF) applied to a signed distance function (SDF) representation. This simple density representation has three benefits: (i) it provides a useful inductive bias to the geometry learned in the neural volume rendering process; (ii) it facilitates a bound on the opacity approximation error, leading to an accurate sampling of the viewing ray. Accurate sampling is important to provide a precise coupling of geometry and radiance; and (iii) it allows efficient unsupervised disentanglement of shape and appearance in volume rendering. Applying this new density representation to challenging scene multiview datasets produced high quality geometry reconstructions, outperforming relevant baselines. Furthermore, switching shape and appearance between scenes is possible due to the disentanglement of the two.

• NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereo, ICCV2021(oral) | [code]

  In this work, we present a new multi-view depth estimation method that utilizes both conventional SfM reconstruction and learning-based priors over the recently proposed neural radiance fields (NeRF). Unlike existing neural network based optimization method that relies on estimated correspondences, our method directly optimizes over implicit volumes, eliminating the challenging step of matching pixels in indoor scenes. The key to our approach is to utilize the learning-based priors to guide the optimization process of NeRF. Our system firstly adapts a monocular depth network over the target scene by finetuning on its sparse SfM reconstruction. Then, we show that the shape-radiance ambiguity of NeRF still exists in indoor environments and propose to address the issue by employing the adapted depth priors to monitor the sampling process of volume rendering. Finally, a per-pixel confidence map acquired by error computation on the rendered image can be used to further improve the depth quality. Experiments show that our proposed framework significantly outperforms state-of-the-art methods on indoor scenes, with surprising findings presented on the effectiveness of correspondence-based optimization and NeRF-based optimization over the adapted depth priors. In addition, we show that the guided optimization scheme does not sacrifice the original synthesis capability of neural radiance fields, improving the rendering quality on both seen and novel views.

• 3D Neural Scene Representations for Visuomotor Control, CoRL2021(oral) | [code]

  Humans have a strong intuitive understanding of the 3D environment around us. The mental model of the physics in our brain applies to objects of different materials and enables us to perform a wide range of manipulation tasks that are far beyond the reach of current robots. In this work, we desire to learn models for dynamic 3D scenes purely from 2D visual observations. Our model combines Neural Radiance Fields (NeRF) and time contrastive learning with an autoencoding framework, which learns viewpoint-invariant 3D-aware scene representations. We show that a dynamics model, constructed over the learned representation space, enables visuomotor control for challenging manipulation tasks involving both rigid bodies and fluids, where the target is specified in a viewpoint different from what the robot operates on. When coupled with an auto-decoding framework, it can even support goal specification from camera viewpoints that are outside the training distribution. We further demonstrate the richness of the learned 3D dynamics model by performing future prediction and novel view synthesis. Finally, we provide detailed ablation studies regarding different system designs and qualitative analysis of the learned representations.

• Vision-Only Robot Navigation in a Neural Radiance World | [code]

  Neural Radiance Fields (NeRFs) have recently emerged as a powerful paradigm for the representation of natural, complex 3D scenes. NeRFs represent continuous volumetric density and RGB values in a neural network, and generate photo-realistic images from unseen camera viewpoints through ray tracing. We propose an algorithm for navigating a robot through a 3D environment represented as a NeRF using only an on-board RGB camera for localization. We assume the NeRF for the scene has been pre-trained offline, and the robot's objective is to navigate through unoccupied space in the NeRF to reach a goal pose. We introduce a trajectory optimization algorithm that avoids collisions with high-density regions in the NeRF based on a discrete time version of differential flatness that is amenable to constraining the robot's full pose and control inputs. We also introduce an optimization based filtering method to estimate 6DoF pose and velocities for the robot in the NeRF given only an onboard RGB camera. We combine the trajectory planner with the pose filter in an online replanning loop to give a vision-based robot navigation pipeline. We present simulation results with a quadrotor robot navigating through a jungle gym environment, the inside of a church, and Stonehenge using only an RGB camera. We also demonstrate an omnidirectional ground robot navigating through the church, requiring it to reorient to fit through the narrow gap. Videos of this work can be found at this https URL .