| CPC H04N 19/119 (2014.11) [H04N 19/132 (2014.11); H04N 19/159 (2014.11); H04N 19/176 (2014.11)] | 20 Claims |

|
1. A method comprising:
determining that multiple shape prediction mode and multi-hypothesis prediction mode are both enabled for an image block;
partitioning the image block into at least a first prediction block and a second prediction block;
deriving a first multiple shape prediction candidate list of at least one motion vector predictor for the first prediction block and a second multiple shape prediction candidate list of at least one motion vector predictor for the second prediction block, the second multiple shape prediction candidate list being different from the first multiple shape prediction candidate list; and
decoding the first and second prediction blocks, wherein decoding the first prediction block comprises obtaining prediction information for the first prediction block based on the first multiple shape prediction candidate list and decoding the second prediction block comprises obtaining prediction information for the second prediction block based on the second multiple shape prediction candidate list;
wherein deriving the first multiple shape prediction candidate list comprises adding at least one intra mode of a first neighboring block to the first multiple shape prediction candidate list or comprises adding at least one intra mode selected in a set of intra modes based on a shape of the first prediction block to the first multiple shape prediction candidate list; and
wherein deriving the second multiple shape prediction candidate list comprises adding at least one intra mode of a second neighboring block to the second multiple shape prediction candidate list on comprises adding at least one intra mode selected in a set of intra modes based on a shape of the second prediction block to the second multiple shape prediction candidate list.
|