Abstract
This paper describe robust approach to track the object, by utilizing visual features. Identical visual features such as color, disparity and shape parameters are used to detect and track the object in unstructured and cluttered environment robustly. An improved active search method is used in order to realize effective tracking. Active search stereo based visual tracking system of humanoid is designed. Two computers are embedded in to the I stein to realize real time robust tracking of object. One computer is used for processing visual information and other one is responsible for visual servo control of pan-tilt platform on which CCD cameras are mounted. The performance of the developed visual cues based tracking of an object is exploited through experiments.