Image coding method based on affine motion prediction, and device for same

A picture encoding method performed by an encoding apparatus, the method comprising:

• generating an affine MVP candidate list for a current block, the affine MVP candidate list including at least one of a first affine MVP candidate block or a second affine MVP candidate block, the generating the affine MVP candidate list comprising:
• selecting, from a first block group, the first affine MVP candidate block that is coded according to an affine motion model, and selecting, from a second block group, the second affine MVP candidate block that is coded according to the affine motion model, wherein the first block group includes (i) a first neighboring block being positioned at a bottom-left corner of a current block and (ii) a second neighboring block being positioned at a left boundary of the current block, and wherein the second block group includes (i) a third neighboring block being positioned at a top-right corner of the current block, (ii) a fourth neighboring block being positioned at a top boundary of the current block, and (iii) a fifth neighboring block being positioned at a top-left corner of the current block;
• selecting one of the first and second affine MVP candidate blocks included in the affine MVP candidate list;
• deriving an affine MVP candidate index representing the selected affine MVP candidate;
• deriving CPMVPs for respective CPs of the current block based on the selected one of the first and second affine MVP candidate blocks;
• deriving CPMVs for the respective CPs of the current block;
• deriving CPMVDs for the respective CPs of the current block based on the CPMVPs and the CPMVs for the respective CPs;
• deriving sub-block motion vectors for the current block based on the CPMVs for the respective CPs of the current block;
• deriving prediction samples for the current block based on the sub-block motion vectors for the current block;
• deriving residual samples for the current block based on the derived prediction samples;
• deriving residual information based on the derived residual samples;
• and
• encoding information related to the affine MVP candidate index, information regarding the derived CPMVDs, and the residual information, wherein the first block group excludes the third, fourth, and fifth neighboring blocks, and wherein the second block group excludes the first and second neighboring blocks.

A non-transitory computer-readable storage medium storing coded information generated by a method, the method comprising:

• generating an affine MVP candidate list for a current block, the affine MVP candidate list including at least one of a first affine MVP candidate block or a second affine MVP candidate block, the generating the affine MVP candidate list comprising:
• selecting, from a first block group, the first affine MVP candidate block that is coded according to an affine motion model, and selecting, from a second block group, the second affine MVP candidate block that is coded according to the affine motion model, wherein the first block group includes (i) a first neighboring block being positioned at a bottom-left corner of a current block and (ii) a second neighboring block being positioned at a left boundary of the current block, and wherein the second block group includes (i) a third neighboring block being positioned at a top-right corner of the current block, (ii) a fourth neighboring block being positioned at a top boundary of the current block, and (iii) a fifth neighboring block being positioned at a top-left corner of the current block;
• selecting one of the first and second affine MVP candidate blocks included in the affine MVP candidate list;
• deriving an affine MVP candidate index representing the selected affine MVP candidate;
• deriving CPMVPs for respective CPs of the current block based on the selected one of the first and second affine MVP candidate blocks;
• deriving CPMVs for the respective CPs of the current block;
• deriving CPMVDs for the respective CPs of the current block based on the CPMVPs and the CPMVs for the respective CPs;
• deriving sub-block motion vectors for the current block based on the CPMVs for the respective CPs of the current block;
• deriving prediction samples for the current block based on the sub-block motion vectors for the current block;
• deriving residual samples for the current block based on the derived prediction samples;
• deriving residual information based on the derived residual samples;
• and
• encoding information related to the affine MVP candidate index, information regarding the derived CPMVDs, and the residual information, wherein the first block group excludes the third, fourth, and fifth neighboring blocks, and wherein the second block group excludes the first and second neighboring blocks.