Methods and apparatuses of determining an alignment level between motion compensated reference patches for reducing motion vector refinement steps are provided. According to one method, obtaining, by a decoder, motion compensated interpolated samples based on sub-pixel accurate merge motion vectors from a bilinear motion compensated interpolation; computing, by the decoder, a sum of absolute differences (SAD) between two motion compensated reference patches using a subset of the motion compensated interpolated samples; determining, by the decoder, whether the SAD is less than a coding unit (CU) size-dependent threshold value; when the SAD is less than the CU size-dependent threshold value: skipping remaining decoder-side motion vector refinement (DMVR) process steps; and performing final motion compensation; and when the SAD is not less than the CU size-dependent threshold value: performing the remaining DMVR process steps; and performing the final motion compensation.

An apparatus for video decoding includes receiving circuitry and processing circuitry. The processing circuitry decodes prediction information for a current block in a current picture from a coded video bitstream, the prediction information being indicative of a prediction mode that is based on an expanded motion vector candidate list. The processing circuitry expands the current block to generate an expanded block by iteratively (i) increasing a width of the current block by a first grid size and (ii) increasing a height of the current block by a second grid size until a number of iterations is equal to a value indicative of a maximum search round. The processing circuitry searches and locates, in the expanded block, a plurality of blocks and constructs, in response to the prediction mode, the expanded motion vector candidate list.

An apparatus for video decoding includes receiving circuitry and processing circuitry. The processing circuitry decodes prediction information for a current block in a current picture from a coded video bitstream, the prediction information being indicative of a prediction mode that is based on an expanded motion vector candidate list. The processing circuitry expands the current block to generate an expanded block by iteratively (i) increasing a width of the current block by a first grid size and (ii) increasing a height of the current block by a second grid size until a number of iterations is equal to a value indicative of a maximum search round. The processing circuitry searches and locates, in the expanded block, a plurality of blocks and constructs, in response to the prediction mode, the expanded motion vector candidate list.

An image coding method includes selecting two or more transform components from among a plurality of transform components that include a translation component and non-translation components, the two or more transform components serving as reference information that represents a reference destination of a current block; coding selection information that identifies the two or more transform components that have been selected from among the plurality of transform components; and coding the reference information of the current block by using reference information of a coded block different from the current block.

An image coding method includes selecting two or more transform components from among a plurality of transform components that include a translation component and non-translation components, the two or more transform components serving as reference information that represents a reference destination of a current block; coding selection information that identifies the two or more transform components that have been selected from among the plurality of transform components; and coding the reference information of the current block by using reference information of a coded block different from the current block.

An encoder which includes circuitry and memory is provided. The circuitry: calculates a cost for each of search points included in a first set; determines whether a base search point has a lowest cost in the first set; when the cost of the base search points is lowest, selects the base search point as a first best search point; when the cost of the base search points is not lowest, calculates a cost for each of search points included in a second set, selects a search point having a lowest cost from among the first set and the second set, as a second best search point, and encodes a current block to be encoded, using a motion vector corresponding to the first best search point or the second best search point.

An encoder which includes circuitry and memory is provided. The circuitry: calculates a cost for each of search points included in a first set; determines whether a base search point has a lowest cost in the first set; when the cost of the base search points is lowest, selects the base search point as a first best search point; when the cost of the base search points is not lowest, calculates a cost for each of search points included in a second set, selects a search point having a lowest cost from among the first set and the second set, as a second best search point, and encodes a current block to be encoded, using a motion vector corresponding to the first best search point or the second best search point.

The present disclosure relates to the iterative fast-search of more than one identified best blocks, corresponding to best patches within a search area, by early termination of the subblock iteration through comparison of a similarity measure to a similarity threshold. In particular, the positions of K-best matched-blocks, divided into multiple subblocks, are found for a reference block within an image search area. By performing subblock-based iterative calculations of the similarity measure between a block at a current position and the reference block within a search area, the iteration progresses as long as the similarity measure value remains larger than a threshold value, and the positions of the K-best patches are recorded in a storage medium, whereas the subblock iteration terminates when the similarity measure value is lower than the threshold value.

The present disclosure relates to the iterative fast-search of more than one identified best blocks, corresponding to best patches within a search area, by early termination of the subblock iteration through comparison of a similarity measure to a similarity threshold. In particular, the positions of K-best matched-blocks, divided into multiple subblocks, are found for a reference block within an image search area. By performing subblock-based iterative calculations of the similarity measure between a block at a current position and the reference block within a search area, the iteration progresses as long as the similarity measure value remains larger than a threshold value, and the positions of the K-best patches are recorded in a storage medium, whereas the subblock iteration terminates when the similarity measure value is lower than the threshold value.

An apparatus for encoding a current frame of a video. The apparatus includes a memory and a processor. The processor is configured to execute instructions stored in the memory to generate, for each reference frame of a subset of available reference frames, at least one respective candidate global motion model (GMM); partition the current frame into blocks; generate an aggregated residual frame for the current frame; and encode the respective residual blocks in a compressed bitstream. To generate the aggregated residual frame includes to select, for predicting each block of the blocks, a respective selected GMM, where the respective selected GMM corresponds to the one of the at least one respective candidate GMMs that minimizes a total error associated with the aggregated residual frame; and obtain respective residual blocks for the block.