Multifeatures Based Compressive Sensing Tracking

To benefit from the development of compressive sensing, we cast tracking as asparse approximation problem in a particle filter framework based on multifeatures. In this framework, the target template is composed of multiplefeatures extracted from visible and infrared frames; in addition, occlusion,interruption, and noises are addressed through a set of trivial templates. Withthis model, the sparsity is achieved via a compressive sensing approach withoutnonnegative constraints; then the residual between sparsity representationand the compressed sensing observation is used to measure the likelihoodwhich weights particles. After that, the target template is adaptively updatedaccording to the Bhattacharyya coefficients. Some experimental resultsdemonstrate that the proposed tracker appears to have better robustness compared withfour different algorithms