Subblock based motion vector predictor displacement vector reordering using template matching

1. A method of video decoding, comprising:

receiving prediction information of a current coding block in a current picture from a coded video bitstream, the prediction information indicating that the current coding block is coded using a subblock-based temporal motion vector prediction (SbTMVP) mode;

obtaining multiple displacement vector (DV) offsets from the coded video bitstream, each DV offset corresponding to a displacement vector candidate;

deriving multiple DV candidates by applying the multiple DV offsets to a fixed DV predictor of the current coding block;

comparing a template of the current coding block with each of multiple templates, each template of the multiple templates being located at a position specified by a corresponding one of the multiple DV candidates;

calculating a cost value associated with each one of the multiple DV candidates based on the comparing;

reordering DV offset indices of the multiple DV candidates based on their calculated cost values; and

predicting the current coding block in the SbTMVP mode based at least on a DV offset index selected from the reordered DV offset indices.

2. The method of claim 1 , further comprising receiving an index signaled in the coded video bitstream, wherein the index indicates which DV offset candidate is selected among the reordered DV offset indices for performing SbTMVP.

3. The method of claim 1 , wherein after the reordering, the method further comprises selecting a DV offset candidate with a lowest calculated template matching cost by default for performing SbTMVP.

4. The method of claim 1 , wherein the cost value is calculated by performing Sum of Absolute Differences (SAD), Sum of Absolute Transformed Differences (SATD), Sum of Squared Error (SSE), sub-sampled SAD, or mean-removed SAD.

5. The method of claim 1 , wherein the multiple DV candidates comprise Merge with Motion Vector Difference (MMVD) candidates.

6. The method of claim 5 , wherein the comparing, the calculating, and the reordering are performed only for a subset of the MMVD candidates, wherein a relative order of one or more other ones of the MMVD candidates is kept unchanged.

7. The method of claim 5 , wherein the comparing, the calculating, and the reordering are performed for all of the MMVD candidates, wherein after reordering only a number N of the MMVD candidates which have a lowest cost are used, wherein the number N is less than or equal to a total number of the MMVD candidates.

8. The method of claim 7 , wherein an index in a range of [0, N−1] is signaled into a bitstream to indicate which MMVD candidate is used, wherein the number N is predefined or is signaled into a high-level syntax.

9. The method of claim 1 , wherein the DV offset indices are reordered by descending or ascending order of cost values calculated for corresponding DV offset candidates.

10. A method of video decoding, comprising:

receiving prediction information of a current coding block in a current picture from a coded video bitstream, the prediction information indicating that the current coding block is coded using a subblock-based temporal motion vector prediction (SbTMVP) mode;

comparing a template of the current coding block with each of multiple templates, each template of the multiple templates being located at a position specified by a corresponding one of multiple displacement vector (DV) predictor candidates;

calculating a cost value associated with each one of the multiple DV predictor candidates based on the comparing;

reordering a list of the multiple DV predictor candidates based on their calculated cost values; and

predicting the current coding block in the SbTMVP mode based at least on a DV predictor selected from the reordered list of DV predictor candidates.

11. The method of claim 10 , further comprising receiving an index signaled in the coded video bitstream, wherein the index indicates which DV predictor candidate is selected from the reordered list of the multiple DV predictor candidates for performing SbTMVP.

12. The method of claim 10 , wherein after the reordering, the method further comprises selecting a DV predictor candidate with a lowest calculated template matching cost by default for performing SbTMVP.

13. The method of claim 10 , wherein the list of the multiple DV predictor candidates is constructed from spatial neighboring coding units (CUs) or from history-based motion vector prediction (HMVP) candidates.

14. The method of claim 10 , wherein after the reordering, only a first N number of the multiple DV predictor candidates on the list are signaled.

15. The method of claim 10 , wherein the multiple DV predictor candidates in the list are reordered by descending or ascending order of their calculated cost values.

16. A method of video decoding, comprising:

receiving prediction information of a current coding block in a current picture from a coded video bitstream, the prediction information indicating that the current coding block is coded using a subblock-based temporal motion vector prediction (SbTMVP) mode;

generating a SbTMVP candidate list including multiple displacement vector (DV) candidates for the current coding block, wherein the SbTMVP candidate list includes at least one DV predictor candidate derived without applying any DV offset and at least one other DV predictor candidate derived by applying a DV offset to a base DV predictor;

comparing a template of the current coding block with each of multiple templates, each template of the multiple templates being located at a position specified by a corresponding one of the multiple DV candidates in the SbTMVP candidate list;

calculating a cost value associated with each one of the multiple DV candidates based on the comparing;

reordering the multiple DV candidates in the SbTMVP candidate list based on their calculated cost values; and

predicting the current coding block in the SbTMVP mode based on a DV predictor candidate selected from the reordered DV candidates in the SbTMVP candidate list.

17. The method of claim 16 , further comprising receiving an index signaled in the coded video bitstream, wherein the index indicates which DV candidate is selected from the reordered SbTMVP candidate list for performing SbTMVP.

18. The method of claim 17 , wherein the selected DV candidate is a SbTMVP DV candidate derived without applying any DV offset or is a SbTMVP Merge with Motion Vector Difference (MMVD) candidate derived by applying a respective DV offset to a respective base DV predictor.

19. The method of claim 18 , wherein, when the selected DV candidate is the SbTMVP MMVD candidate, a signaling of the respective base DV predictor and a signaling of an index of the respective DV offset are performed separately.

20. The method of claim 16 , wherein the SbTMVP candidate list is constructed independently of an affine merge candidate list when template matching based reordering of candidates is enabled for a current frame, wherein when a subblock merge mode is signaled, an additional syntax is signaled at coding block level to indicate whether to construct the SbTMVP candidate list or the affine merge candidate list.