| CPC H04N 19/521 (2014.11) [H04N 19/105 (2014.11); H04N 19/139 (2014.11); H04N 19/176 (2014.11); H04N 19/537 (2014.11)] | 4 Claims |
|
1. A video decoding method, the method comprising:
decoding merge mode based motion vector refine usage information;
deriving an initial motion vector, based on an initial motion vector index information indicating a merge candidate used as the initial motion vector for deriving a refined motion vector, from a merge candidate list of the current block;
decoding motion compensation information of the current block indicating a motion vector difference;
deriving a refined motion vector indicating a position in a cross-shape region based on the motion compensation information; and
deriving a prediction block of the current block based on the refined motion vector,
wherein the cross-shape region is based on a position indicated by the initial motion vector,
wherein the deriving the initial motion vector includes, when the initial motion vector index information is not decoded, deriving a motion vector of a merge candidate corresponding to a predetermined number of the merge candidate list as the initial motion vector.
|