Objective: An new algorithm to estimation the scale and orientation of the tracking window is presented in order to solve the problem that traditional Mean Shift based tracking algorithm often fail when tracking a deformable target because of the unchanged tracking window. Method: Firstly
the principal components of the variance matrix are adopted to compute the scale and orientation of tracking target. Further
the similarity measure and Kalman filter are used to update the tracking window. Result: The experimental results show that the algorithm can be implemented in real-time and can adapt to the changes of scale and orientation of the moving object. Conclusion: It can satisfy the requirements of tracking non-rigid objects.