Method and apparatus of improved intra luma prediction mode coding utilizing block size of neighboring blocks

摘要

A method of video coding using intra prediction in which the intra prediction modes are ranked according to a priority order associated with the block size. One or more tables are used, where the tables ranks the intra prediction modes according to a first priority order for the block having a first block size and ranks the intra prediction modes according to a second priority order for the block having a second block size. Two or more neighboring intra prediction modes corresponding to two or more neighboring blocks are received, where each neighboring block has a neighboring block size corresponding to the first block size or the second block size. A highest-priority mode among said two or more neighboring intra prediction modes is selected as the most probable mode. The current mode is then encoded or decoded using the most probable mode as a predictor.

主权项

1. A method of video coding using intra prediction for a block of video data, the method comprising:
determining one or more tables ranking first intra prediction modes according to a first priority order for a first neighboring block having a first block size and second intra prediction modes according to a second priority order for a second neighboring block having a second block size, wherein the first block size is different from the second block size, and wherein the first priority order is different from the second priority order; receiving a current intra prediction mode corresponding to a current block; receiving two or more neighboring intra prediction modes corresponding to two or more neighboring blocks, wherein one of said two or more neighboring block has a neighboring block size corresponding to the first block size and the other one of said two or more neighboring block has a neighboring block size corresponding to the second block size; determining a mode priority for each of said two or more neighboring intra prediction modes according to each neighboring intra prediction mode and each neighboring block size of said two or more neighboring blocks using said one or more tables; determining a highest-priority mode among said two or more neighboring intra prediction modes as a most probable mode by identifying a neighboring intra prediction mode with a lowest index relative to all of the two or more neighboring intra prediction modes ranked in their respective tables corresponding to different block sizes; and encoding the current mode using the most probable mode as a predictor.