/Distilled-Light-GaitSet

A lightweight network for efficient gait recognition. A distillation framework for efficient training.

Primary LanguagePython

Distilled Light GaitSet

Dataset & Preparation

Download CASIA-B Dataset

Download Parameter files of GaitSet into work-distilled/teacher/GaitSet

Before training, please make sure you have prepared the dataset and the parameter files of GaitSet

The silhouettes after pretreatment MUST have a size of 64x64

Futhermore, you also can test our code on OU-MVLP Dataset

Citation

Please cite these papers in your publications if it helps your research:

@article{song2022distilled,
  title={Distilled light GaitSet: Towards scalable gait recognition},
  author={Song, Xu and Huang, Yan and Huang, Yan and Shan, Caifeng and Wang, Jilong and Chen, Yu},
  journal={Pattern Recognition Letters},
  volume={157},
  pages={27--34},
  year={2022},
  publisher={Elsevier}
}

Link to paper: