Skip to content
Implementation of "Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields" for AI challenger keypoint competition
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.ipynb_checkpoints
config
cython
pose_io
sample_image
README.md
TrainWeight.py
evaluate.py
evaluation_Pose_testImage_easy.ipynb
evaluation_Pose_trainImage_easy.ipynb
generateLabelCPMWeight.py
modelCPMWeight.py
test_generatelabel.ipynb

README.md

Implementation of "Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields" for AI challenger keypoint competition

Train demo

  1. install cython package
./cython/rebuild.sh

  1. Generate intermediate files

change folder name and json name in pose_io/parse_label.py

path1 = '/data/guest_users/liangdong/liangdong/practice_demo/AIchallenger/keypoint_validation_annotations_20170911.json' 
trainimagepath = '/data/guest_users/liangdong/liangdong/practice_demo/AIchallenger/validation_image/keypoint_validation_images_20170911/'
python pose_io/parse_label.py 
  1. Train
python TrainWeight.py

You can download mxnet model and parameters for vgg19 from here

Cite paper Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields

@article{cao2016realtime,
  title={Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields},
  author={Zhe Cao and Tomas Simon and Shih-En Wei and Yaser Sheikh},
  journal={arXiv preprint arXiv:1611.08050},
  year={2016}
  }

Other implementations of Realtime Multi-Person 2D Pose Estimation

Original caffe training model

Original data preparation and demo

Pytorch

keras

mxnet

You can’t perform that action at this time.