Method for encoding/decoding a video using a merge mode with prediction unit partitioning and a bitstream storing instructions to transmit compressed video data generated by the method of encoding the video
A video decoding method, according to the present invention, comprises the steps of: deriving an L0 motion vector and an L1 motion vector of a current block; deriving an L0 prediction sample and an L1 prediction sample for a first position inside the current block on the basis of the L0 motion vector and the L1 motion vector; determining whether to apply a bidirectional optical flow to the current block; deriving a refined motion vector for a subblock which comprises the first position, when it is determined to apply the bidirectional optical flow; deriving a refined prediction sample for the first position on the basis of the refined motion vector; and obtaining a prediction sample for the first position by using the L0 prediction sample, the L1 prediction sample, and the refined prediction sample.
1. A method of decoding a video, the method comprising:
determining whether a merge mode with prediction unit (PU) partitioning is applied to a current block or not,
wherein in response that the merge mode with PU partitioning is applied to the current block, the method further comprises:
determining a partition type of the current block, the current block being divided into a first partition and a second partition;
obtaining first motion information for the first partition based on a first merge candidate included in a merge candidate list of the current block; and
obtaining second motion information for the second partition based on a second merge candidate included in the merge candidate list of the current block,
wherein the first merge candidate is selected from the merge candidate list based on a first merge index decoded from a bitstream,
wherein the second merge candidate is selected from the merge candidate list based on a second merge index decoded from the bitstream,
wherein the merge mode with PU partitioning is allowed for the current block only when a width-height ratio of the current block and a height-width ratio of the current block are less than a threshold value,
wherein the partition type of the current block is selected among a plurality of partition type candidates, and
wherein the plurality of partition type candidates includes a first partition type candidate in which a size of the first partition and a size of the second partition are mutually the same and a second partition type candidate in which the size of the first partition and the size of the second partition are different from the each other.
2. The method of claim 1 , wherein whether the second merge index is decoded from the bitstream or not is determined based on a maximum number of the merge candidates.
3. The method of claim 2 , wherein the maximum number of the merge candidates is determined based on information indicating a maximum number difference between when the merge mode with PU partitioning is not applied and when the merge mode with PU partitioning is applied.
4. The method of claim 3 , wherein when the maximum number of the merge candidates is less than a threshold value, decoding the second merge index from the bitstream is omitted.
5. The method of claim 4 , wherein when decoding the second merge index from the bitstream is omitted, a value of the second merge index is inferred to the default value, the default value being 0.
6. The method of claim 1 , wherein the partition type of the current block is selected among a plurality of partition type candidate and
wherein one of the partition type candidates, in which directions of each of partitioning lines are the same but positions of each of the partitioning lines are different, is selected based on index information specifying one of the positions.
7. A method of encoding a video, the method comprising:
determining whether a merge mode with prediction unit (PU) partitioning is applied to a current block or not,
wherein in response that the merge mode with PU partitioning is applied to the current block, the method further comprises:
determining a partition type of the current block, the current block being divided into a first partition and a second partition;
obtaining first motion information for the first partition based on a first merge candidate included in a merge candidate list of the current block; and
obtaining second motion information for the second partition based on a second merge candidate included in the merge candidate list of the current block,
wherein a first merge index specifying the first merge candidate is encoded into a bitstream,
wherein a second merge index specifying the second merge candidate is encoded into the bitstream, and
wherein the merge mode with PU partitioning is allowed for the current block only when a width-height ratio of the current block and a height-width ratio of the current block are less than a threshold value,
wherein the partition type of the current block is selected among a plurality of partition type candidates, and
wherein the plurality of partition type candidates includes a first partition type candidate in which a size of the first partition and a size of the second partition are mutually the same and a second partition type candidate in which the size of the first partition and the size of the second partition are different from the each other.
8. A non-transitory computer readable recording medium for storing instructions which when executed by a processor cause the processor to transmit compressed video data, the compressed video data being generated by an encoding method which comprising:
determining whether a merge mode with prediction unit (PU) partitioning is applied to a current block or not,
wherein in response that the merge mode with PU partitioning is applied to the current block, the method further comprises:
determining a partition type of the current block, the current block being divided into a first partition and a second partition;
obtaining first motion information for the first partition based on a first merge candidate included in a merge candidate list of the current block; and
obtaining second motion information for the second partition based on a second merge candidate included in the merge candidate list of the current block,
wherein a first merge index specifying the first merge candidate is encoded into a bitstream,
wherein a second merge index specifying the second merge candidate is encoded into the bitstream, and
wherein the merge mode with PU partitioning is allowed for the current block only when a width-height ratio of the current block and a height-width ratio of the current block are less than a threshold value,
wherein the partition type of the current block is selected among a plurality of partition type candidates, and
wherein the plurality of partition type candidates includes a first partition type candidate in which a size of the first partition and a size of the second partition are mutually the same and a second partition type candidate in which the size of the first partition and the size of the second partition are different from the each other.
9. The method of claim 1 , wherein the second merge candidate is determined by comparing the second merge index with the first merge index,
wherein in response to the second merge index is less than the first merge index, a merge candidate whose index is identical to the second merge index is determined as the second merge candidate, and
wherein in response to the second merge index is equal to or greater than the first merge index, a merge candidate whose index is identical to a value adding 1 to the second merge index is determined as the second merge candidate.
10. The method of claim 1 , wherein when the first merge candidate has both L0 motion information and L1 motion information, one of the L0 motion information and the L1 motion information of the first merge candidate is derived as the first motion information.
11. The method of claim 10 , wherein selection of one of the L0 motion information and the L1 motion information of the first merge candidate is based on whether an index of the first merge candidate is odd number or even number.