RPG-Palm

RPG-Palm Realistic Pseudo-data Generation for Palmprint Recognition | Paper | Arxiv

Pytorch implementation for RPG-Palm . Our improved work PCE-Palm can be found in link

Example results

Prerequisites

• Python 3
• CPU or NVIDIA GPU + CUDA CuDNN

Getting Started

This code borrows heavily from the BicycleGAN repository. You can find more details about the original code in the BicycleGAN repository.

Installation

• Clone this repo:

git clone https://github.com/Ukuer/rpg-palm.git
cd rpg-palm

• Install PyTorch and dependencies from http://pytorch.org
• Install python libraries visdom, dominate, and moviepy.
• Noted that do not install vlkit from pypi, because it is not compatible with the current version. Please install vlkit as following:

pip install palm-recognition/vlkit-0.1.0b11.tar.gz 

Use a Pre-trained Model

• Download checkpoints and place it in ./checkpoints .

• Then bash ./run_rpg.sh

• Noted that you should modify some contents in ./run_rpg.sh to meet you requirements.

Model Training

• To train a model,

bash train.sh

• To view training results and loss plots, run python -m visdom.server and click the URL http://localhost:8097. To see more intermediate results, check out ./checkpoints/NAME/web/index.html

Citation

If you find this useful for your research, please use the following.

@inproceedings{shen2023rpg,
  title={RPG-Palm: Realistic Pseudo-data Generation for Palmprint Recognition},
  author={Shen, Lei and Jin, Jianlong and Zhang, Ruixin and Li, Huaen and Zhao, Kai and Zhang, Yingyi and Zhang, Jingyun and Ding, Shouhong and Zhao, Yang and Jia, Wei},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
  pages={19605--19616},
  year={2023}
}

@inproceedings{jin2024pce,
  title={PCE-Palm: Palm Crease Energy Based Two-Stage Realistic Pseudo-Palmprint Generation},
  author={Jin, Jianlong and Shen, Lei and Zhang, Ruixin and Zhao, Chenglong and Jin, Ge and Zhang, Jingyun and Ding, Shouhong and Zhao, Yang and Jia, Wei},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={38},
  number={3},
  pages={2616--2624},
  year={2024}
}

If you have any questions or encounter any issues with the RPG-PALM code, please feel free to contact me (email: [email protected]). I would be more than happy to assist you in any way I can.

Acknowledgements

This code borrows heavily from the BicycleGAN repository.