Physics-Based Models for Human Motion Analysis


Image Group Talk by David J Fleet, Department of Computer Science, University of Toronto


This talk concerns the use of physics-based models for human pose tracking and scene inference. We outline our motivation for physics-based models, some results with monocular pose tracking in terms of biomechanically inspired controllers, and recent results on the inference of scene interactions.

We show that physics-based models facilitate the estimation of physically plausible human motion with little or no mocap data required. Scene interactions play an integral role in modeling sources of external forces acting on the body.

More info:

This is joint work with Marcus Brubaker and Leonid Sigal.

Scientific Host: Kim Steenstrup Pedersen