Skip to content
View kxqt's full-sized avatar
  • PKU
  • Beijing, China

Highlights

  • Pro

Block or report kxqt

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse

Pinned Loading

  1. Expedit-LargeScale-Vision-Transformer/Expedit-SAM Expedit-LargeScale-Vision-Transformer/Expedit-SAM Public

    [NeurIPS2022] This is the official implementation of the paper "Expediting Large-Scale Vision Transformer for Dense Prediction without Fine-tuning" for SAM.

    Jupyter Notebook 82 8

  2. RockeyCoss/Prompt-Segment-Anything RockeyCoss/Prompt-Segment-Anything Public

    This is an implementation of zero-shot instance segmentation using Segment Anything.

    Python 301 15

  3. Expedit-LargeScale-Vision-Transformer/Expedit-DINO Expedit-LargeScale-Vision-Transformer/Expedit-DINO Public

    [NeurIPS2022] This is the official implementation of the paper "Expediting Large-Scale Vision Transformer for Dense Prediction without Fine-tuning" on DINO and MaskDINO.

    Python 8 1

  4. H-Deformable-DETR H-Deformable-DETR Public

    Forked from HDETR/H-Deformable-DETR

    [CVPR2023] This is the official implementation of H-Deformable-DETR w/ ViT-L (MAE).

    Python 6

  5. Expedit-LargeScale-Vision-Transformer/Expedit-DPT Expedit-LargeScale-Vision-Transformer/Expedit-DPT Public

    [NeurIPS2022] This is the official implementation of the paper "Expediting Large-Scale Vision Transformer for Dense Prediction without Fine-tuning" on DPT.

    Python 10

  6. Expedit-LargeScale-Vision-Transformer/Expedit-Segmenter Expedit-LargeScale-Vision-Transformer/Expedit-Segmenter Public

    [NeurIPS2022] This is the official implementation of the paper "Expediting Large-Scale Vision Transformer for Dense Prediction without Fine-tuning" on Segmenter.

    Jupyter Notebook 6