MMVD signaling improvement

1. A method for video decoding in a video decoder, comprising:

receiving prediction information of a plurality of blocks in a current picture from a coded video bitstream;

determining, based on the prediction information, that at least one of a decoder-side motion vector refinement (DMVR) mode or a bi-directional optical flow (BDOF) mode is allowed for the plurality of blocks;

in response to the determination that the at least one of the DMVR mode or the BDOF mode is allowed for the plurality of blocks, determining that a plurality of merge with motion vector difference (MMVD) merge flags is inferred as false to indicate that an MMVD mode is not applied for the plurality of blocks; and

in response to the determination that the plurality of MMVD merge flags is inferred as false for the plurality of blocks, reconstructing each block in the plurality of blocks without applying the MMVD mode.

2. The method of claim 1 , wherein the prediction information includes at least one flag indicating whether the at least one of the DMVR mode or the BDOF mode is allowed for the plurality of blocks.

3. The method of claim 2 , further comprising:

determining that (i) the current picture refers to a forward reference picture and a backward reference picture, and (ii) the current picture is in the middle of the forward reference picture and the backward reference picture;

determining that the plurality of MMVD merge flags is not signaled; and

inferring the plurality of MMVD merge flags based on a determination that the MMVD mode is not allowed in the current picture.

4. The method of claim 3 , wherein the determination that the current picture is in the middle of the forward reference picture and the backward reference picture includes determining that an absolute value of a first picture order count (POC) difference between the forward reference picture and the current picture is equal to an absolute value of a second POC difference between the backward reference picture and the current picture.

5. The method of claim 4 , wherein the absolute value of the first POC difference and the absolute value of the second POC difference are less than a threshold.

6. The method of claim 1 , wherein

the prediction information includes a MMVD merge enabling flag of the plurality of blocks,

the MMVD merge enabling flag indicates that the MMVD mode is not enabled for the plurality of blocks, and

the method includes determining that the plurality MMVD merge flags of the plurality of blocks is inferred as false.

7. The method of claim 1 , wherein the plurality of blocks is a slice in the current picture.

8. The method of claim 1 , wherein

in response to a determination that the plurality of MMVD merge flags is not inferred as false for the plurality of blocks, a respective MMVD merge flag of each block in the plurality of blocks is signaled in the coded video bitstream, and

the method further comprises:

selecting a context among a first context and a second context based on whether one or more of the DMVR mode or the BDOF mode is allowed for the plurality of blocks, and

decoding the respective MMVD merge flag of each block in the plurality of blocks using context-adaptive binary arithmetic coding (CABAC) with the selected context.

9. The method of claim 8 , wherein

the prediction information indicates whether conditions to select the first context are met,

the conditions include:

the one or more of the DMVR mode or the BDOF mode is allowed for the plurality of blocks,

reference pictures of the plurality of blocks include a forward reference picture and a backward reference picture, and

an absolute value of a first picture order count (POC) difference between the forward reference picture and the current picture is equal to an absolute value of a second POC difference between the backward reference picture and the current picture, and the selecting the context includes:

selecting the first context as the context based on the conditions being satisfied; and

selecting the second context as the context based on one of the conditions not being satisfied.

10. The method of claim 9 , wherein

the first context is initialized with a probability indicating that the MMVD merge flag of each block in the plurality of blocks has a probability to be false that is higher than a probability that the MMVD merge flag is true.

11. An apparatus for video decoding, comprising:

processing circuitry configured to:

receive prediction information of a plurality of blocks in a current picture from a coded video bitstream;

determine, based on the prediction information, that at least one of a decoder-side motion vector refinement (DMVR) mode or a bi-directional optical flow (BDOF) mode is allowed for the plurality of blocks;

in response to the determination that the at least one of the DMVR mode or the BDOF mode is allowed for the plurality of blocks, determine that a plurality of merge with motion vector difference (MMVD) merge flags is inferred as false to indicate that an MMVD mode is not applied for the plurality of blocks; and

in response to the determination that the plurality of MMVD merge flags is inferred as false for the plurality of blocks, reconstruct each block in the plurality of blocks without applying the MMVD mode.

12. The apparatus of claim 11 , wherein the prediction information includes at least one flag indicating whether the at least one of the DMVR mode or the BDOF mode is allowed for the plurality of blocks.

13. The apparatus of claim 12 , wherein the processing circuitry is configured to:

determine that (i) the current picture refers to a forward reference picture and a backward reference picture, and (ii) the current picture is in the middle of the forward reference picture and the backward reference picture;

determine that the plurality of MMVD merge flags is not signaled; and

infer the plurality of MMVD merge flags based on a determination that the MMVD mode is not allowed in the current picture.

14. The apparatus of claim 13 , wherein the processing circuitry is configured to:

determine that an absolute value of a first picture order count (POC) difference between the forward reference picture and the current picture is equal to an absolute value of a second POC difference between the backward reference picture and the current picture.

15. The apparatus of claim 11 , wherein

the prediction information includes a MMVD merge enabling flag of the plurality of blocks,

the MMVD merge enabling flag indicates that the MMVD mode is not enabled for the plurality of blocks, and

the processing circuitry is configured to determine that the plurality MMVD merge flags of the plurality of blocks is inferred as false.

16. The apparatus of claim 11 , wherein the plurality of blocks is a slice in the current picture.

17. The apparatus of claim 11 , wherein

in response to a determination that the plurality of MMVD merge flags is not inferred as false for the plurality of blocks, a respective MMVD merge flag of each block in the plurality of blocks is signaled in the coded video bitstream, and

the processing circuitry is configured to:

select a context among a first context and a second context based on whether one or more of the DMVR mode or the BDOF mode is allowed for the plurality of blocks, and

decode the respective MMVD merge flag of each block in the plurality of blocks using context-adaptive binary arithmetic coding (CABAC) with the selected context.

18. The apparatus of claim 17 , wherein

the prediction information indicates whether conditions to select the first context are met, the conditions include:

the one or more of the DMVR mode or the BDOF mode is allowed for the plurality of blocks,

reference pictures of the plurality of blocks include a forward reference picture and a backward reference picture, and

an absolute value of a first picture order count (POC) difference between the forward reference picture and the current picture is equal to an absolute value of a second POC difference between the backward reference picture and the current picture, and the processing circuitry is configured to:

select the first context as the context based on the conditions being satisfied; and

select the second context as the context based on one of the conditions not being satisfied.

19. The apparatus of claim 18 , wherein

the first context is initialized with a probability indicating that the MMVD merge flag of each block in the plurality of blocks has a probability to be false that is higher than a probability that the MMVD merge flag is true.

20. A non-transitory computer-readable storage medium storing a program executable by at least one processor to perform:

receiving prediction information of a plurality of blocks in a current picture from a coded video bitstream;

determining, based on the prediction information, that at least one of a decoder-side motion vector refinement (DMVR) mode or a bi-directional optical flow (BDOF) mode is allowed for the plurality of blocks;

in response to the determination that the at least one of the DMVR mode or the BDOF mode is allowed for the plurality of blocks, determining that a plurality of merge with motion vector difference (MMVD) merge flags is inferred as false to indicate that an MMVD mode is not applied for the plurality of blocks; and

in response to the determination that the plurality of MMVD merge flags is inferred as false for the plurality of blocks, reconstructing each block in the plurality of blocks without applying the MMVD mode.