Date: 09/02/2018


Micha Livne
Department of Computer Science
University of Toronto
Toronto, Canada

Leonid Sigal
Department of Computer Science
University of British Columbia
Vancouver, Canada

Marcus A. Brubaker
Lassonde School of Engineering
York University
Toronto, Canada

David J. Fleet
Micha Livne (
Department of Computer Science
University of Toronto
Toronto, Canada


Was accepted as an oral talk in 2018 15th Conference on Computer and Robot Vision (CRV).


We propose a new generative approach to physics-based motion capture. Unlike prior attempts to incorporate physics into tracking, which assume the subject and scene geometry are calibrated and known a priori, our approach is automatic and online. This distinction is important since calibration of the environment is often difficult, especially for motions with props, uneven surfaces, or outdoor scenes. The use of physics in this context provides a natural framework to reason about contact and the plausibility of recovered motions. We propose a fast data-driven parametric body model, based on linear-blend skinning, which decouples deformations due to pose, anthropometrics and body shape. This model facilitates estimation of body proportions and leverages these in the physics and for improving fitting. Pose (and shape) parameters are estimated using robust ICP optimization with physics-based dynamic priors that incorporate contact. Contact is estimated from torque trajectories and predictions of which contact points were active. To our knowledge, this is the first approach to take physics into account without explicit a priori knowledge of the environment or body dimensions. We demonstrate effective tracking from a noisy single depth camera, improving on state-of-the-art results quantitatively and producing better qualitative results, reducing visual artifacts like foot-skate and jitter.



Results Summary

Please let the video time to load. The video below has lower quality for the sake of responsiveness. See above for higher quality version.

More Results

Body Mesh Model



Body Mesh Model


Registration and Tracking