| CPC H04N 19/105 (2014.11) [H04N 19/137 (2014.11); H04N 19/176 (2014.11)] | 18 Claims |
|
1. A method for video decoding, comprising:
partitioning a picture into a plurality of coding units;
partitioning a coding unit of the plurality of coding units into two geometric partitions;
constructing a merge list for either of the two geometric partitions; and
performing a predetermined number of motion vector comparison operations, the motion vector comparison operation comprising comparison between a current motion vector and a motion vector that is already selected in the merge list during a process of constructing the merge list,
wherein constructing the merge list comprises:
constructing a first merge list comprising a plurality of candidates based on a merge list construction process for regular merge prediction, wherein each of the plurality of candidates comprises a List 0 motion vector or a List 1 motion vector or both the List 0 motion vector and the List 1 motion vector; and
obtaining motion vectors for either of the two geometric partitions by selecting the List 0 motion vector or the List 1 motion vector of the plurality of candidates based on merge indexes of the plurality of candidates,
wherein the method further comprises:
in response to determining that the predetermined number is equal to one, obtaining a first motion vector from a first candidate in the first merge list by:
selecting a List 0 motion vector of the first candidate into the merge list, in response to determining that the List 0 motion vector of the first candidate is available; and
selecting a List 1 motion vector of the first candidate into the merge list, in response to determining that the List 0 motion vector of the first candidate is unavailable.
|