Skip to content
/ PCRP Public
forked from LZU-SIAT/PCRP

Prototypical Contrast and Reverse Prediction: Unsupervised Skeleton based Action Recognition

License

Notifications You must be signed in to change notification settings

Mikexu007/PCRP

 
 

Repository files navigation

PCRP

Introduction

This is the official implementation of "Prototypical Contrast and Reverse Prediction: Unsupervised Skeleton based Action Recognition".

Requirements

  • Python 3.6
  • Pytorch 1.0.1

Datasets

Put the data into the folder that matches the codes in pc_test.py

Usage

  • pretrain and then linear evaluation:
    python pc_test.py

License

PCRP is released under the MIT License.

Citation

@misc{xu2020prototypical,
title={Prototypical Contrast and Reverse Prediction: Unsupervised Skeleton Based Action Recognition},
author={Shihao Xu and Haocong Rao and Xiping Hu and Bin Hu},
year={2020},
eprint={2011.07236},
archivePrefix={arXiv},
primaryClass={cs.CV}
}

About

Prototypical Contrast and Reverse Prediction: Unsupervised Skeleton based Action Recognition

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%