Blocking artifact reduction of DCT coded image sequences using a visually adaptive postprocessing

The blocking effect is known to be the major degradation in block based video compression techniques. In this paper, we describe an adaptive postprocessing algorithm for the reduction of blocking effect in video coded sequences. It is based on a visual model for the prediction of blockiness visibility. The postfiltering operation is based on a 3D adaptive filter. A space-variant spatial filter is first used to smooth the pixels at block boundaries where the blocking effect is highly visible. It is followed by motion compensated nonlinear filtering in the temporal domain. Experimental results are presented showing that the proposed algorithm can remove the blocking effect while keeping image sharpness.