OpenUnGait is a comprehensive and efficient codebase designed to facilitate research in Unsupervised Gait Recognition. It offers a robust framework for Unsupervised Domain Adaptation (UDA), enabling researchers to easily train, evaluate, and benchmark methods that transfer knowledge from labeled source domains to unlabeled target environments across diverse scenarios
The following algorithms and datasets are currently supported. More methods and benchmarks will be continuously integrated.
Please see 0.get_started.md.
- 🛠️ Complete training and evaluation code
- 📦 Release pre-trained model checkpoints
@InProceedings{Ma2023fine,
author = {Ma, Kang and Fu, Ying and Zheng, Dezhi and Peng, Yunjie and Cao, Chunshui and Huang, Yongzhen},
title = {Fine-grained Unsupervised Domain Adaptation for Gait Recognition},
booktitle = ICCV,
month = {October},
year = {2023},
pages = {11313-11322}
}Our code refers to the following repositories.
We thank the authors for releasing their codes