Decoder-side control point motion vector refinement for affine inter-prediction in video coding

A device for decoding video data, the device comprising:

• a memory configured to store video data;
• and
• a processing system comprising one or more processors implemented in circuitry, the processing system being configured to:
• perform a first decoder-side motion vector refinement (DMVR) process to refine a first control point motion vector (CPMV) of a current block of the video data to form a first refined CPMV for the current block;
• perform, independent of the performance of the first DMVR process, a second DMVR process to refine a second CPMV of the current block of video data to form a second refined CPMV for the current block, the second DMVR process being distinct from the first DMVR process;
• form a prediction block for the current block using the first refined CPMV and the second refined CPMV according to affine prediction mode;
• and
• decode the current block using the prediction block.

A non-transitory, computer-readable storage medium having stored thereon instructions that, when executed, cause a processor to:

• perform a first decoder-side motion vector refinement (DMVR) process to refine a first control point motion vector (CPMV) of a current block of video data to form a first refined CPMV for the current block;
• perform, independent of the performance of the first DMVR process, a second DMVR process to refine a second CPMV of the current block of video data to form a second refined CPMV for the current block, the second DMVR process being distinct from the first DMVR process;
• form a prediction block for the current block using the first refined CPMV and the second refined CPMV according to affine prediction mode;
• and
• decode the current block using the prediction block.