This repository offers a synthetic point cloud dataset with ground truth skeletons of multiple species.
Below are some images of point clouds from the dataset:
You can download the data by following this link.
Our dataset uses subtractive augmentations based on Perlin Noise.
We offer a GPU-based implementation of Perlin Noise, available in this repository.
For data visualization, we recommend our previously developed library, accessible via this link.
An example augmentation can be done using python augmentation.py
Please use the following BibTeX entry to cite our work:
@article{dobbs2024quantifying,
title={Quantifying robustness: 3D tree point cloud skeletonization with smart-tree in noisy domains},
author={Dobbs, Harry and Batchelor, Oliver and Peat, Casey and Atlas, James and Green, Richard},
journal={Pattern Analysis and Applications},
volume={27},
number={1},
pages={28},
year={2024},
publisher={Springer}
}
Should you have any questions, comments or suggestions please use the following contact details: [email protected]