Skip to content

Latest commit

 

History

History
28 lines (23 loc) · 841 Bytes

README.md

File metadata and controls

28 lines (23 loc) · 841 Bytes

limb-analysis

Setup

  1. Set up passwordless ssh login to vishal@10.237.23.241 (or any server where AlphaPose is installed). Follow instructions at https://linuxize.com/post/how-to-setup-passwordless-ssh-login/
  2. Install opencv in a python 3.8 environment
conda create -n env python=3.8
conda activate env
pip install opencv-python

Usage

  1. Comparing angles at shoulder and elbow in 2 videos
python3 compare_vids.py <path to video file 1> <path to video file 2>
  1. Obtaining joint data from AlphaPose
python3 get_data.py <path to video file>

Results will be saved in results-<video name> directory in .json format, as given at https://github.com/MVIG-SJTU/AlphaPose/blob/master/docs/output.md

  1. Obtaining angles at shoulder and elbow from json
python3 check_angles.py <path to json file>