Simultaneous motion estimation and filtering of image sequences

In this paper, me present an algorithm to simultaneously estimate multiple- frame motions and filter image sequences, The relative motion d(k)(x) betwe en the reference frame s(x) and the kth frame s(x - d(k)(x)) in the presenc e of additive white Gaussian noise (AWGN) is estimated using the maximum-li kelihood (ML) principle. The reference frame is also filtered in the linear minimum mean square error (LMMSE) sense during the process of motion estim ation. Simulation experiments are performed using the affine motion model t o illustrate the performance of this method.