Iterative technique for 3-D motion estimation in videophone applications

In object based coding of facial images, the accuracy of motion and depth parameter estimates strongly affects the coding efficiency. We propose an improved algorithm based on stochastic relaxation for 3-D motion and depth estimation that converges to true motion and depth parameters even in the presence of 50% error in the initial depth estimates. The proposed method is compared with an existing algorithm (MBASIC) in case of different number of point correspondences. The simulation results show that the proposed method provides significantly better results than the MBASIC algorithm.