Video signal processing method and device
An image decoding method, according to the present invention, can comprise the steps of: deriving a spatial merge candidate of a current block; generating a merge candidate list for the current block on the basis of the spatial merge candidate; acquiring motion information on the current block on the basis of the merge candidate list; and performing motion compensation for the current block by using the motion information.
1. A method of decoding a video, the method comprising:
determining whether to divide a coding block into two partitions or not;
when it is determined to divide the coding block, dividing the coding block into a first partition and a second partition;
deriving spatial merge candidates from spatial neighboring blocks, the spatial neighboring blocks being determined based on a position and a size of the coding block;
generating a merge candidate list comprising the spatial merge candidates, the merge candidate list being shared for the first partition and the second partition;
obtaining motion information of the first partition based on a first merge index;
obtaining motion information of the second partition based on a second merge index; and
performing motion compensation for the first partition and the second partition based on their respective motion information,
wherein the first merge index is used for specifying one of merge candidates included in the merge candidate list, and
wherein the second merge index is used for specifying one of merge candidates excluding a specified one by the first merge index.
2. The method of claim 1 , wherein the merge candidate list is generated by adding the spatial merge candidates and a temporal merge candidate thereto, and
wherein if a number of merge candidates added to the merge candidate list is less than a threshold value, an average merge candidate generated based on two among the merge candidates is further added to the merge candidate list.
3. The method of claim 2 , wherein if a first merge candidate among the two merge candidates has both L0 and L1 motion information while a second merge candidate among the two merge candidates has only L0 motion information, an L0 motion vector of the average merge candidate is derived by averaging L0 motion vectors of the first merge candidate and the second merge candidate while an L1 motion vector of the average merge candidate is derived to be the same as an L1 motion vector of the first merge candidate.
4. The method of claim 1 , wherein the merge candidate list is generated by adding the spatial merge candidates and a temporal merge candidate thereto, and
wherein if a number of merge candidates added to the merge candidate list is less than a threshold value, a spatial merge candidate derived from a spatial block that has a different location with the spatial neighboring blocks and is included in the same CTU (Coding Tree Unit) with the coding block is newly added to the merge candidate list.
5. The method of claim 4 , wherein the spatial neighboring blocks comprises a top-neighboring block, a left neighboring block, a left-bottom neighboring block, a top-right neighboring block and a left-top neighboring block.
6. A method of encoding an image, the method comprising:
determining whether to divide a coding block into two partitions or not;
when it is determined to divide the coding block, dividing the coding block into a first partition and a second partition;
deriving spatial merge candidates from spatial neighboring blocks, the spatial neighboring blocks being determined based on a position and a size of the coding block;
generating a merge candidate list comprising the spatial merge candidates, the merge candidate list being shared for the first partition and the second partition;
obtaining motion information of the first partition based on a first merge candidate in the merge candidate list;
obtaining motion information of the second partition based on a second merge candidate the merge candidate list; and
performing motion compensation for the first partition and the second partition based on their respective motion information,
wherein the first merge index used for specifying the first merge candidate among merge candidates included in the merge candidate list is encoded in a bitstream, and
wherein the second merge index used for specifying the second merge candidate among merge candidates excluding the first merge candidate is encoded in a bitstream.
7. The method of claim 6 , wherein the merge candidate list is generated by adding the spatial merge candidates and a temporal merge candidate thereto, and
wherein if a number of merge candidates added to the merge candidate list is less than a threshold value, an average merge candidate generated based on two among the merge candidates is further added to the merge candidate list.
8. The method of claim 7 , wherein if a first one among the two merge candidates has both L0 and L1 motion information while a second one candidate among the two merge candidates has only L0 motion information, an L0 motion vector of the average merge candidate is derived by averaging L0 motion vectors of the first one and the second one while an L1 motion vector of the average merge candidate is derived to be the same as an L1 motion vector of the first one.
9. The method of claim 6 , wherein the merge candidate list is generated by adding the spatial merge candidates and a temporal merge candidate thereto, and
wherein if a number of merge candidates added to the merge candidate list is less than a threshold value, a spatial merge candidate derived from a spatial block that has a different location with the spatial neighboring blocks and is included in the same CTU (Coding Tree Unit) with the coding block is newly added to the merge candidate list.
10. The method of claim 9 , wherein the spatial neighboring blocks comprises a top-neighboring block, a left neighboring block, a left-bottom neighboring block, a top-right neighboring block and a left-top neighboring block.
11. A non-transitory computer-readable medium for storing data associated with a video signal, comprising:
a data stream stored in the non-transitory computer-readable medium, the data stream being encoded by an encoding method which comprising:
determining whether to divide a coding block into two partitions or not;
when it is determined to divide the coding block, dividing the coding block into a first partition and a second partition;
deriving spatial merge candidates from spatial neighboring blocks, the spatial neighboring blocks being determined based on a position and a size of the coding block;
generating a merge candidate list comprising the spatial merge candidates, the merge candidate list being shared for the first partition and the second partition;
obtaining motion information of the first partition based on a first merge candidate in the merge candidate list;
obtaining motion information of the second partition based on a second merge candidate the merge candidate list; and
performing motion compensation for the first partition and the second partition based on their respective motion information,
wherein the first merge index used for specifying the first merge candidate among merge candidates included in the merge candidate list is encoded in a bitstream, and
wherein the second merge index used for specifying the second merge candidate among merge candidates excluding the first merge candidate is encoded in a bitstream.