In this paper we propose a robust object tracking algorithm which includes appearance model and motion model based on sparse representation. As the main challenge for object tracking is to account for drastic appearance change, we propose a robust appearance model that exploits both holistic templates and local representations. We develop a sparsity-based discriminative classifier and a sparsity-based generative model. In the sparsity-based discriminative classifier module, we introduce an effective method to compute the confidence value that assigns more weights to the foreground than the background. In the sparsity-based generative model, we propose a novel histogram-based method that takes the spatial information of each patch into consideration with an occlusion handing scheme. We fuse the two modules reasonably and propose a robust collaborative appearance model for object tracking. Furthermore, the proposed update scheme considers both the latest observations and the original template, thereby enabling the tracker to deal with appearance change effectively and alleviate the drift problem. As for the motion model, we improve the original motion model in order to promote the tracking efficiency and assure the tracking accuracy. Numerous experiments on various challenging videos demonstrate that the proposed tracker performs favorably against several state-of-the-art algorithms. |