Review and improvement areas of mean shift tracking algorithm

The goal of this article is to review the state-of-the-art object tracking techniques developed using mean shift approach and identify the domain in which it can be improved upon. Object tracking, in general, is a challenging problem, which can be achieved in many ways. Of the various possible directions, mean shift is most popular because of its simplicity and applicability to many states of affairs. The tracking algorithm developed upon the mean shift platform can be investigated in either of the two focal points; firstly, for accurate object centroid detection and secondly, for scale adaptation. Present paper tries to investigate the approaches reported in both the domains and identifies the improvement areas in respective field.