Motion compensation encoding apparatus and motion compensation encoding method for high-efficiency encoding of video information through selective use of previously derived motion vectors in place of motion vectors derived from motion estimation

摘要

In a video motion compensation encoding apparatus in which each of periodically selected pictures is encoded as successive blocks of picture elements, by encoding of the prediction error amounts obtained for each block in conjunction with variable-length encoding by a motion vector encoder of motion vectors which are utilized in motion compensation processing to derive the prediction error amounts, each of these motion vectors is selected by a comparator-controlled switch either as an estimated motion vector derived by a conventional type of motion estimation section or as a predicted motion vector derived by a motion vector prediction section based on previously encoded motion vectors, with the estimated motion vector being selected only if it is found that an amount of prediction error which would occur from using the predicted motion vector is significantly greater than that which would result from using the estimated motion vector. A higher probability of encoding successively identical motion vectors is thereby achieved, resulting in greater encoding efficiency.