Multiband width Kernel-Based Object Tracking

Object tracking using Mean Shift (MS) has been attracting considerable attention recently. In this paper, we try to deal withone of its shortcoming. Mean shift is designed to find local maxima for tracking objects. Therefore, in large target movementbetween two consecutive frames, the local and global modes are not the same as previous frames so that Mean Shift trackermay fail in tracking the desired object via localizing the global mode. To overcome this problem, a multibandwidth procedure isproposed to help conventional MS tracker reach the global mode of the density function using any staring points. This graduallysmoothening procedure is called Multi Bandwidth Mean Shift (MBMS) which in fact smoothens the Kernel Function through amultiple kernel-based sampling procedure automatically. Since i t is important for us to have less computational complexity forreal-time applications, we try to decrease the number of iterations to reach the global mode. Based on our results, this proposedversion of MS enables us to track an object with the same initial point much faster than conventional MS tracker.