Video Processing Motion Data Refinement for Coding Efficiency

The patent applies preliminary action by performing motion data refinement before motion compensation. The motion data of coded blocks is refined in advance based on neighboring block information and prediction data, so that when motion compensation is subsequently performed, the prediction precision is already optimized, thereby improving overall coding efficiency without increasing computational complexity during the critical compensation stage.

The patent implements feedback by using the refined motion data from neighboring blocks to improve the motion compensation process. The motion data of a current block is refined based on the motion data of neighboring blocks and prediction data, creating a feedback loop where the prediction results are used to further improve the motion estimation, thereby enhancing coding efficiency iteratively.

The patent performs motion data refinement as a preliminary action before motion compensation. By pre-refining the motion data of all coded blocks in a picture based on neighboring block information and prediction data, the system optimizes prediction precision in advance. This approach distributes the computational load before the critical motion compensation stage, avoiding time consumption during the compensation process itself.

The patent changes the parameter of motion data precision by refining motion vectors to higher precision levels. The motion data is refined based on neighboring block motion data and prediction data, allowing the system to achieve higher prediction precision without requiring a complete re-estimation of motion vectors, thus managing processing time effectively.