Quadtree structured region-wise motion compensation for video compression

The conventional variable-size block motion-compensation technique, even th ough superior to the fixed-size block motion-compensation technique, cannot fully utilize the motion information of a frame for its partitioning. This paper presents a quadtree structured region-wise motion-compensation techn ique that utilizes more effectively the motion content of a frame in terms of the shape, size, and location of the partitioned regions. The proposed t echnique is based on a new coding scheme of the quadtree structure, where a two-bit code is used. The partitioning of a given frame is carried out thr ough the processes of merging and combining. The merging process partitions the frame into a number of regions by carrying out all possible 4-to-1, 3- to-1, and 2-to-1 merges along the quadtree hierarchy according to some spec ified criteria,while the combining process combines some of the partitioned regions that have the same motion vector into a single region. The motion vectors of the resulting partitioned regions are coded by a mew coding sche me that uses the temporal correlation of the motion fields. Test results of applying the proposed method on a number of MPEG video sequences are inclu ded. These results indicate that the proposed method can provide a signific antly improved rate-distortion performance.