https://github.com/gengshan-y/dyn_pose

Dynamic pose estimation in MXNet. faster-rcnn detection + cpm pose estimation + LSTM dynamical modeling
https://github.com/gengshan-y/dyn_pose

Last synced: 2 months ago
JSON representation

Dynamic pose estimation in MXNet. faster-rcnn detection + cpm pose estimation + LSTM dynamical modeling

• Host: GitHub
• URL: https://github.com/gengshan-y/dyn_pose
• Owner: gengshan-y
• License: mit
• Created: 2017-05-20T03:00:09.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2017-12-14T19:07:58.000Z (about 7 years ago)
• Last Synced: 2024-08-01T22:41:14.847Z (5 months ago)
• Language: Jupyter Notebook
• Homepage:
• Size: 1.16 MB
• Stars: 5
• Watchers: 3
• Forks: 1
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: License.md

Awesome Lists containing this project

• Awesome-MXNet - Dynamic pose estimation

README

        
# dyn_pose

Training and inference code for [LSTM-based action recognition model](http://www.contrib.andrew.cmu.edu/~gengshay/wordpress/index.php/424-2/).

Online version to be released later.

## Steps

- [models](https://drive.google.com/open?id=1v1TeDwCWb426g_LWnuKkVHr8PZZR-P2S)

- pre-process data: preprocess.py / split.py

- extract pose: extract.py

- train model: train_lstm.py

- evaluate model: infer_lstm.py

- offline demo: example.py

## Notes

- modify label_num in model/lstm.config

## Utils

- display.ipynb: display pose estimation results

- run.sh: evaluation script

- test.ipynb: miscellaneous

## TODO

- reduce spatial net size

## Reference 

- [Convolutional Pose Machines](https://github.com/shihenw/convolutional-pose-machines-release)