Adaptive Learning of Statistical Appearance Models for 3D Human Tracking
Timothy J. Roberts, Stephen James McKenna, Ian W. Ricketts · 2002
A likelihood formulation for human tracking is presented based upon matching feature statistics on the surface of an articulated 3D model. A benefit of such a formulation over current techniques is that it provides a dense, object-based cue. Multi-dimensional histograms are used to represent the feature distributions. Different histogram similarity measures are evaluated for tracking purposes. An on-line region grouping algorithm, driven by prior knowledge of clothing structure, is derived that improves the appearance estimation and computational efficiency. Finally, we demonstrate that the smooth, broad likelihood response allows efficient inference using coarse sampling and local optimisation. Results from tracking real world sequences are presented.