Mission

poseinterface exists to advance pose estimation and point tracking applications in animal behaviour videos. The project aims to:

• Build PoseMark, an open benchmark corpus featuring dozens of annotated videos from multiple institutes, with a contribution model that enables participation from the wider community.

• Develop a general-purpose Python framework, provided through the poseinterface Python package, for running pose estimation and point-tracking tools on PoseMark, and for benchmarking their outputs using standardised evaluation metrics.

• Provide baseline models trained on PoseMark, using common pose estimation and tracking frameworks.