Visual object tracking via sparse reconstruction
Li Li, Zhenjun Han, Jianbin Jiao, Qixiang Ye · 2010
This paper proposes a new approach based on sparse representation for visual object tracking. The sparse representation is implemented by exploiting L1--norm minimization, which most compactly expresses the object and rejects all other possible but less compact representations. With the coefficient vector of the sparse representation, we reconstruct the tracked object in an instantaneous sample set, which improves the tracking adaptation to background variation, object shape change, and partial occlusion. Our experiments on public datasets show state-of-the-art results, which are better than those of several representative tracking methods.