Kernel covariance image region description for object tracking

We propose a nonlinear covariance region descriptor for target tracking. The target object appearance and spatial information is represented using a covariance matrix in a target derived Hilbert space using kernel principal component analysis. A similarity measure is derived, which computes the similarity of a candidate image region to the learned covariance matrix. A variational technique is provided to maximize the similarity measure, which iteratively finds the best matched object region. Tracking performance is demonstrated on a variety of sequences containing noise, occlusions, illumination changes, background clutter, etc.