SPU-Net: Self-Supervised Point Cloud Upsampling by Coarse-to-Fine Reconstruction with Self-Projection Optimization

Created by Xinhai Liu, Zhizhong Han, Yu-Shen Liu, Matthias Zwicker.

Usage

For training, the cmd example:

    python train_spu-net.py --gpu gpu_index --log_dir log/spu_00012_git

For evaluating, the cmd example:

    python evaluate.py --gt data/test/groundtruth/ --pred log/spu_00012_git/output/

The training data from PU-GAN:

path to the test point cloud and ground-truths: data/test

You can download the entire dataset from PU-GAN or Google Drive.

Citation

If you find our work useful in your research, please consider citing:

    @article{liu2022spu,
      title={SPU-Net: Self-Supervised Point Cloud Upsampling by Coarse-to-Fine Reconstruction with Self-Projection Optimization},
      author={Liu, Xinhai and Liu, Xinchen and Liu, Yu-Shen and Han, Zhizhong},
      journal={IEEE Transactions on Image Processing},
      volume={31},
      pages={4213--4226},
      year={2022},
      publisher={IEEE}
    }