Human tracking system plays a critical role in many applications such as surveillance and robot
applications. Our earlier implementation [1] while quite successful was restricted to applications with
approximately fixed camera. In this paper we present some recent work that removes this restriction. Such
systems are required for machine vision from moving platforms such as robots, intelligent vehicles and
unattended large field of regard cameras with a small field of view. Our approach is based on the use of a
modified motion estimation technique coupled with our previous work in [1] for human tracking. This
allows introducing an invariant human tracking system that is robust to background and lighting changes in
addition to shadows remo val, handles unknown camera/ human motion and does not require special
hardware components or camera calibrations.ochemistry