An adaptive motion-compensated filter using cumulants

Abstract

This paper deals with the use of higher order statistics (HOS) in a spatio-temporal filtering scheme to reduce noise in video sequences. Prior to filtering, motion compensation is performed by a robust estimator. Then, we apply a 3-D order statistic filter. Both stages are based on Kurtosis minimization. Experimental results on noisy real video sequences illustrate the efficiency of the proposed approach.