METHOD AND APPARATUS FOR MOTION COMPENSATION WITH NON-SQUARE SUB-BLOCKS IN VIDEO CODING

摘要

A method, apparatus and computer program product select a sub-block shape and size for high-order motion compensation for the purpose of utilizing the selected sub-blocks in the motion compensation process in a video encoding or decoding process. The method, apparatus and computer program product receive a bitstream comprising one or more coding units. For each coding unit, the method, apparatus and computer program product determine if the coding unit is to be motion compensated by using a translational motion model or a higher order motion model. In response to determining that the coding unit is to be motion compensated by using a translational motion model or a higher order motion model, the method, apparatus and computer program product determine a dimension of the coding unit; determine a shape and size for a set of sub-blocks in the coding unit; and perform motion compensated prediction for the set of sub-blocks.