Intra-frame compression of video sequences with Inverse Difference Pyramid decomposition

The paper presents one new method for intra frame compression of video sequences based on the inverse difference pyramid decomposition. The compression starts with the processing of a selected frame in a group of consecutive TV frames. The coarse approximation of this frame is calculated and it is then used as a reference for the whole group. The remaining TV frames in the group are processed sequentially one by one, using the coarse approximation of the reference frame and correcting this approximation in correspondence to the currently processed one. The individual additional information, which represents the details for every frame is added to the coded video sequence data. The new method for intra frame processing offers much higher compression than M-JPEG for same quality. When compared to MPEG the method ensures better quality for the fast moving objects.