Skip to main content

DeepLabCut

DeepLabCut provides markerless pose estimation for tracking animal body parts in videos and images. Relay supports both inference with pre-trained models and training custom models.

Input formats

.mp4, .avi, .mov, .mkv, .tiff, .png, .jpg

Pre-trained models

Parameters

Outputs

Presets

Compute requirements

Training custom models

You can train custom DeepLabCut models on your own labeled data:

Base architectures

Training parameters

Training presets

Training outputs

  • Trained model (ZIP)
  • Checkpoint files (.pb)
  • Loss curves (JSON + PNG)
  • Evaluation metrics (JSON)