Unsupervised Distribution-aware Keypoints Generation from 3D Point Clouds

Description

This repository contains the code for our paper: Unsupervised Distribution-aware Keypoints Generation from 3D Point Clouds.

Unsupervised Distribution-aware Keypoints Generation from 3D Point Clouds,
Yiqi Wu, Xingye Chen, Xuan Huang, Kelin Song, Dejun Zhang
Bibetex

Environment setup

pip install torch==1.8.1+cu111 torchvision==0.9.1+cu111 torchaudio==0.8.1 -f https://download.pytorch.org/whl/torch_stable.html
pip install pointnet2_ops_lib/.

Dataset

The training and testing data for correspondence is provided by KeypointNet and ShapeNet

Citation

@article{wu2024unsupervised,
  title={Unsupervised distribution-aware keypoints generation from 3D point clouds},
  author={Wu, Yiqi and Chen, Xingye and Huang, Xuan and Song, Kelin and Zhang, Dejun},
  journal={Neural Networks},
  pages={106158},
  year={2024},
  publisher={Elsevier}
}

Acknowledgment

Our implementation is mainly based on the following codebases. We gratefully thank the authors for their wonderful works.

3DStructurePoints, Pointnet2_PyTorch