xiachenrui/kpg-rl

Keypoint-Guided Optimal Transport

Keypoint-Guided Optimal Transport

Official codes for "Keypoint-Guided Optimal Transport with Applications in Heterogeneous Domain Adaptation".

Requirements

python3.6
scipy==1.7.1
numpy==1.20.3
matplotlib==3.4.3
cvxpy==1.2.0
pot==0.8.1.0
pytorch==1.5.0

Instructions

The codes for HDA experiments (Section 5.2), Open-Set HDA experiments (Section 5.3), UDA experiments (Appendix B.3), toy experiments for KPG-RL (Section 5.1), and toy experiments for Partial KPG-RL (Appendix B.1) are respectively in folders "HDA", "Open-Set HDA", "UDA", "toy_data", and "toy_data_partial". Please follow the Readme.md in each folder to run the codes.

Citation:

@inproceedings{
gu2022keypointguided,
title={Keypoint-Guided Optimal Transport with Applications in Heterogeneous Domain Adaptation},
author={Xiang Gu and Yucheng Yang and Wei Zeng and Jian Sun and Zongben Xu},
booktitle={Advances in Neural Information Processing Systems},
editor={Alice H. Oh and Alekh Agarwal and Danielle Belgrave and Kyunghyun Cho},
year={2022},
url={https://openreview.net/forum?id=m6DJxSuKuqF}
}

Contact

For any problem, please do not hesitate to contact [email protected].

The code is based on code of POT packadge.