出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：Recently, visual tracking based on sparse principle component analysis has drawn muchresearch attention. As we all know, principle component analysis (PCA) is widely used in dataprocessing and dimensionality reduction. But PCA is difficult to interpret in practicalapplication and all those principal components are linear combinations of all variables. In ourpaper, a novel visual tracking method based on sparse principal component analysis and L1tracking is introduced, which we named the method SPCA-L1 tracking. We firstly introducetrivial templates of L1 tracking method, which are used to describe noise, into PCA appearancemodel. Then we use lasso model to achieve sparse coefficients. Then we update the eigenbasisand mean incrementally to make the method robust when solving different kinds of changes ofthe target. Numerous experiments, where the targets undergo large changes in pose, scale andillumination, demonstrate the effectiveness and robustness of the proposed method.
关键词：Visual tracking; sparse principal component analysis; particle filter
