Partial occlusions are commonplace in a variety of real world computer vision applications: surveillance, intelligent environments, assistive robotics, autonomous navigation, etc. While occlusion handling methods have been proposed, most methods tend to break down when confronted with numerous occluders in a scene. In this project, we are developing layered image-plane representations for tracking through substantial occlusions. An image-plane representation of motion around an object is associated with a pre-computed graphical model, which can be instantiated efficiently during online tracking. A global state and observation space can be obtained by linking transitions between layers. A Reversible Jump Markov Chain Monte Carlo approach can then be used to infer the number of people and track them online. Given the instantiated models, a Reversible Jump Markov Chain Monte Carlo approach can then be used to infer the number of people and track them online.

The above figure illustrates the basic idea behind these layered graphical models. Occluders' masks are shown in (a), while bitmaps corresponding to observation regions in the image plane are shown in (b). These bitmaps are placed to roughly capture motion of a person walking around the corresponding occluding object (c). Transitions between layers (not shown in this figure) are explained in the paper.