MatConvNet_Realtime_Multi-Person_Pose_Estimation

This is a matlab implementation of Realtime Multi-Person Pose Estimation using matconvnet backend, origin code is here https://github.com/ZheC/Realtime_Multi-Person_Pose_Estimation

Introduction

Code repo for personal reproducing 2017 CVPR Oral paper using MatConvNet.

This project is licensed under the terms of the license.

Disclaimer

Copyright is owned by the [original author](Zhe Cao).

Require

• MatConvNet

cd <Pose_matconvnet>
git clone https://github.com/vlfeat/matconvnet.git
cd matconvnet
run matlab/vl_compilenn ;

• COCO Model (The model is freely available for free non-commercial use.)

  Download from google drive or BaiduYun

  or

  Convert caffe models to matconvnet by yourself.

  • Download caffemodel

    cd <Pose_matconvnet>
    cd models
    wget http://posefs1.perception.cs.cmu.edu/Users/ZheCao/pose_iter_440000.caffemodel
    

  • MATLAB

    cd models
    Converter_caffe_matconvnet
    build_openpose_net
    

Testing

• MATLAB

cd <Pose_matconvnet>
cd testing
demo_matconvnet_pose();

Related repository

• Our new C++ library openPose
• CVPR'16, Convolutional Pose Machines
• Pytorch version of the code

Citation

Please cite the paper in your publications if it helps your research:

@inproceedings{cao2017realtime,
  author = {Zhe Cao and Tomas Simon and Shih-En Wei and Yaser Sheikh},
  booktitle = {CVPR},
  title = {Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields},
  year = {2017}
  }
  
@inproceedings{wei2016cpm,
  author = {Shih-En Wei and Varun Ramakrishna and Takeo Kanade and Yaser Sheikh},
  booktitle = {CVPR},
  title = {Convolutional pose machines},
  year = {2016}
  }