In order to overcome the disadvantages of just adopting single feature and lacking of local structure information in traditional tracking algorithms,a hybrid model visual tracking algorithm based on multi-feature fusion is proposed. Firstly,it structures a robust appearance model via integrating the local appearance model of intensity together with the global templates of color histograms and histogram of oriented gradient ( HOG) . Then,a strategy of outlier rejection is also proposed,which divides the sparse coefficient into two collaborative components and imposes the l2,1 mixed-norm regularization. The experimental results on benchmark dataset show that the proposed method is more accurate and robust in dealing with illumination change and occlusion.
NETL
NSTL
万方数据
维普
