OpenPose Training

Contents

1. Introduction
2. Functionality
3. Testing
4. Training
5. Citation
6. License

Introduction

OpenPose Training includes the training code for OpenPose, as well as some experimental models that might not necessarily end up in OpenPose (to avoid confusing its users with too many models).

It is authored by Gines Hidalgo, Zhe Cao, Yaadhav Raaj, Tomas Simon, Haroon Idrees, Donglai Xiang, Shih-En Wei, Hanbyul Joo, and Yaser Sheikh. It is based on the papers described in the Citation section and in Realtime Multi-Person Pose Estimation. In addition, OpenPose would not be possible without the CMU Panoptic Studio dataset. We would also like to thank all the people who helped OpenPose in any way.

This repository and its documentation assumes knowledge of OpenPose. If you have not used OpenPose yet, you must familiare yourself with it before attempting to follow this documentation.

Functionality

• Training code for OpenPose.
• Release of some experimental models that have not been included into OpenPose. These models are experimental and might present some issues compared to the models officially released inside OpenPose. This project is licensed under the terms of the license.
  • BODY_135: Whole-body pose estimation models from Single-Network Whole-Body Pose Estimation.
  • BODY_25B: Alternative to the BODY_25 model of OpenPose, with higher accuracy but slower speed.

Experimental Models

The experimental_models directory contains our experimental models, including the whole-body model from Single-Network Whole-Body Pose Estimation, as well as instructions to make it run inside OpenPose. See experimental_models/README.md for more details.

Testing

See testing/README.md for more details.

Training

The training/ directory contains multiple scripts to generate the scripts for training and to actually train the models. See training/README.md for more details.

Validation

The validation/ directory contains multiple scripts to evaluate the accuracy of the trained models. See validation/README.md for more details.

Citation

Please cite these papers in your publications if it helps your research (the face keypoint detector was trained using the procedure described in [Simon et al. 2017] for hands):

@inproceedings{hidalgo2019singlenetwork,
  author = {Gines Hidalgo and Yaadhav Raaj and Haroon Idrees and Donglai Xiang and Hanbyul Joo and Tomas Simon and Yaser Sheikh},
  booktitle = {ICCV},
  title = {Single-Network Whole-Body Pose Estimation},
  year = {2019}
}

@inproceedings{cao2018openpose,
  author = {Zhe Cao and Gines Hidalgo and Tomas Simon and Shih-En Wei and Yaser Sheikh},
  booktitle = {arXiv preprint arXiv:1812.08008},
  title = {Open{P}ose: realtime multi-person 2{D} pose estimation using {P}art {A}ffinity {F}ields},
  year = {2018}
}

Links to the papers:

• Single-Network Whole-Body Pose Estimation
• OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields

License

OpenPose is freely available for free non-commercial use, and may be redistributed under these conditions. Please, see the license for further details. Interested in a commercial license? Check this FlintBox link. For commercial queries, use the Directly Contact Organization section from the FlintBox link and also send a copy of that message to Yaser Sheikh.