3-D non-rigid motion estimation from image sequence based on Makov random field

We propose an approach to 3-D non-rigid motion estimation from image sequence in this paper. First, with the establishment of feature point correspondence between consecutive image frames, the affine motion model and the central projection model are presented for local non-rigid motion. Then, in order to obtain the global motion parameters and overcome the ill-posed 3-D estimation problem, a framework of Markov random field (MRF) is proposed. By incorporating the motion prior constrains into the MRF, the motion smoothness feature-between local regions is reflected. This converts the ill-posed problem into a well-posed one and guarantees a robust solution. Experimental results from a sequence of synthetic image sequence demonstrate the feasibility of the proposed approach.