This repo contains a browser-based GUI that facilitates the deveopment of a pose estimation project.
The app allows you to:
- label data (upload videos, extract frames, annotate keypoints)
- train and evaluate models
- run inference on new videos
Additionally, the app comes with an example dataset if you want to explore, without needing to label your own data.
As of June 2024, Lightning Pose is now published in Nature Methods!
Please see the YouTube tutorial series or Pose app documentation for instructions on how to install and run the app.
Learn more about the core algorithm powering the app via the Lightning Pose github or Lightning Pose documentation.
The Lightning Pose app is primarily maintained by Matt Whiteway (Columbia University), Shmuel Shomrat (Columbia University), and Dan Biderman (Stanford University).
The Lightning Pose app is under active development and we welcome community contributions (see contributing guidelines here). Please get in touch with us on Discord.