Video coding combining intra-sub partition and template-based intra-mode derivation techniques

A video decoding device comprising:

• a processor configured to:
• obtain a coding block that comprises a plurality of sub-partitions;
• for a first sub-partition of the plurality of sub-partitions, derive a first intra prediction mode based on a first template associated with the first sub-partition;
• decode the first sub-partition based on the first intra prediction mode;
• for a second sub-partition of the plurality of sub-partitions, deriving a second intra prediction mode based on a second template associated with the second sub-partition;
• and
• decode the second sub-partition based on the second intra prediction mode.

A video decoding method comprising:

• obtaining a coding block that comprises a plurality of sub-partitions;
• for a first sub-partition of the plurality of sub-partitions, deriving a first intra prediction mode based on a first template associated with the first sub-partition;
• decoding the first sub-partition based on the first intra prediction mode;
• for a second sub-partition of the plurality of sub-partitions, deriving a second intra prediction mode based on a second template associated with the second sub-partition;
• and
• decoding the second sub-partition based on the second intra prediction mode.

The method of claim 9, further comprising:

• obtaining a plurality of probable prediction modes associated with the coding block;
• determining a plurality of predictions of the coding block based on the plurality of probable prediction modes associated with the coding block;
• for the first sub-partition, deriving a secondary first intra prediction mode based on the plurality of predictions of the coding block, wherein decoding the first sub-partition is further based on the secondary first intra prediction mode;
• and
• for the second sub-partition, deriving a secondary second intra prediction mode based on the plurality of predictions of the coding block, wherein decoding the second sub-partition is further based on the secondary second intra prediction mode.

A video encoding device comprising:

• a processor configured to:
• obtain a coding block that comprises a plurality of sub-partitions;
• for a first sub-partition of the plurality of sub-partitions, derive a first intra prediction mode based on a first template associated with the first sub-partition;
• encode the first sub-partition based on the first intra prediction mode;
• for a second sub-partition of the plurality of sub-partitions, deriving a second intra prediction mode based on a second template associated with the second sub-partition;
• and
• encode the second sub-partition based on the second intra prediction mode.