In this paper we describe our work on 3-D model-based tracking of unconstrained human upper body movement. Using real image sequences acquired from multiple views, we recover the 3-D body pose at each time instant without the use of markers. The pose-recovery problem is formulated as a search problem and entails finding the pose parameters of a graphical human model for which its synthesized appearance is most similar to the actual appearance of the real human in the multi-view images. We use a decomposition approach and a best-first technique to search through the high dimensional pose parameter space. Chamfer matching is used as a fast similarity measure between synthesized and real edge images. We illustrate our approach on real data acquired simultaneously from three views.
展开▼