6.1 Introduction

Intelligent visual surveillance is an important application area for computer vision. In situations where networks of hundreds of cameras are used to cover a wide area, the obvious limitation becomes the users' ability to manage such vast amounts of information. For this reason, automated tools that can generalise about activities or track objects are important to the operator. Key to the users' requirements is the ability to track objects across (spatially separated) camera scenes. However, extensive geometric knowledge about the site and camera position is typically required. Such an explicit mapping from camera to world is infeasible for large installations as it requires that the operator know which camera to switch to when an object disappears. To further compound the problem the installation costs of CCTV systems outweigh those of the hardware. This means that geometric constraints or any form of calibration (such as that which might be used with epipolar constraints) is simply not realistic for a real world installation. The algorithms cannot afford to dictate to the installer. This work attempts to address this problem and outlines a method to allow objects to be related and tracked across cameras without any explicit calibration, be it geometric or colour.

Algorithms for tracking multiple objects through occlusion normally perform well for relatively simple scenes where occlusions by static objects and perspective effects are not severe. For example, Figure 6.1 shows a typical surveillance camera view with two distinct regions A and B...