Deep Point Set Resampling via Gradient Fields

Paper code

The official code repository for our TPAMI paper 'Deep Point Set Resampling via Gradient Fields'.

Installation

You can install via conda environment .yaml file

conda env create -f environment.yml
conda activate deeprs

Dataset

to release soon

Training and Testing

To train or test your model, you should first edit the corresponding config file according to the comments (fill in the file directory). Then, you can simply run

## Train a network for denoising
python -m scripts.train_denoise --config ./configs/denoise.yml
## Train a network for upsampling
python -m scripts.train_upsample --config ./configs/ups.yml

for training, and run

## Save your denoising results
python -m scripts.save_denoised --config ./configs/denoise.yml [--ckpt]
## Save your upsampling results
python -m scripts.train_upsample --config ./configs/ups.yml [--ckpt_model] [--ckpt_gen]

for testing.

Evaluation

You can run

python -m scripts.evaluate --config ./configs/denoise.yml
python -m scripts.evaluate_upsample --config ./configs/ups.yml

to evaluate your results

Citation

If you feel this work helpful, please cite

@ARTICLE{9775211,
  author={Chen, Haolan and Du, Bi'an and Luo, Shitong and Hu, Wei},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
  title={Deep Point Set Resampling via Gradient Fields}, 
  year={2022},
  volume={},
  number={},
  pages={1-1},
  doi={10.1109/TPAMI.2022.3175183}}

and contact [email protected] for any question.