This is the official repository for the publication iPUNet: Iterative Cross Field Guided Point Cloud Upsampling.
If you should encounter any problems with the code, don't hesitate to contact me at guangshunwei@gmail.com.
git clone https://github.com/GSW-D/iPUNet.git
cd iPUNet-main
conda env create -f iPUNet.yml
conda activate iPUNet
- [chamfer_distance]
- [KNN]
- [PointNet2] custom ops must be built by running python setup.py install under pointnet2_utils/custom_ops folder.
Dataset can be downloaded from Google Drive.
python train.py
python test.py
@article{wei2023ipunet,
title={iPUNet: Iterative Cross Field Guided Point Cloud Upsampling},
author={Wei, Guangshun and Pan, Hao and Zhuang, Shaojie and Zhou, Yuanfeng and Li, Changjian},
journal={IEEE Transactions on Visualization and Computer Graphics},
year={2023},
publisher={IEEE}
}