Region covariance descriptor based probabilistic object tracking using enhanced similarity criterion

In this paper, a new similarity criterion approach to compare region covariance descriptors (RCD) is proposed. In object tracking, RCDs constructed using feature points of the object exhibit low performance. It is aimed that the performance of the RCD based object tracking approach increases by using proposed method. Object tracking is performed via particle filter and full search approaches. It can also be seen from the experimental results that the proposed method outperforms the conventional RCD based tracking approach.