ABSTRACT

The application of this project is to track a number of people even in crowded natural environments. A system of this sort would be useful in a number of applications, such as human-computer interaction, surveillance, and mobile robots that work among people. The robustness of the system will be able to operate even with partial occlusions of the subjects, and recover from tracking errors where a subject is temporarily lost. Finally, the techniques will be implemented on standard available hardware, and run fast enough to track in real time mobile system. It can be used to assist users in tracking colleagues and friends within a campus environment.