Abstract: A probabilistic method for tracking 3D articulated human figures in image sequences is presented. Within a Bayesian probabilistic framework we learn statistical models of the appearance of humans and general scenes and exploit these models for tracking complex, deformable, human motion. A prior probability distribution over possible human motions is learned from 3D motion-capture data and is combined with learned image likelihood measures for Bayesian tracking using particle filtering. This prior term exploits ideas from texture synthesis to construct implicit probabilistic models of human motion that replace the problem of representation with that of efficient search. By combining multiple image cues, by using learned likelihood models, and by using learned prior models of motion, we demonstrate the tracking of people in monocular image sequences with cluttered scenes and a moving camera.

Joint work with:
  Hedvig Sidenbladh (Swedish Defense Research Agency (FOI))
  David Fleet (PARC Inc.)
  Leonid Sigal (Brown University)