🙋‍♀️ DeepLabCut™ is an efficient method for 2D and 3D markerless pose estimation based on transfer learning with deep neural networks.

• The main code repository is here. Contains code for all necessary steps to do pose estimation incl. GUI.
• 👩‍💻 Read the documentation for how to get started and how to use DeepLabCut.

We also have several related code repositories that may be of interest:

• DLC-Live! & DLC-Live! GUI -- export your trained DeepLabCut model and run it for live analysis. Consider using our easy GUI to do so.
• DLCUtils -- this repo hosts some code and key links to helper packages that input the outputs of DeepLabCut for downstream analysis.
• DLC Workshop & Educational Materials -- learn about how to use and develop with DeepLabCut!
• DLC2Kinematics -- use this code to load DeepLabCut (H5) files for movement analysis, dev. by MW Mathis Lab.
• DLC2Action -- for action segmentation based on pose estimation data, dev. by A Mathis Group.
• CEBRA -- use this code to load DeepLabCut (H5) files for advnced nonlinear embedding analysis and joint modeling, dev. by MW Mathis Lab.

🌈 Contribution guidelines to this project can be found here!

🍿 Find more resources, our papers, and examples on the main website!

🧙 Check out the repos below to find helper code, additional tools, and more!