Abstract: A visual object tracking algorithm is presented using a spatial color model and particle filtering. For lack of spatial information, the observer likelihood based on the color histogram often leads to inaccurate localization when the background has a similar color to that of the objects. The spatial-color histogram which combines the color and spatial information is used to model the object. By relying on the Bhattacharyya distance of the spatial color histogram, a new observer likelihood is built and applied to particle filtering. Experimental results show that the proposed algorithm can achieve more accurate and robust tracking results.

Key words: visual object tracking, particle filter, spatial color histogram