Merge Mode With Motion Vector Difference Based Subblock-Based Temporal Motion Vector Prediction

A method for coding a current block, comprising:

• partitioning the current block into sub-blocks;
• identifying a first neighboring block and a second neighboring block of the current block;
• and
• for each sub-block of at least some of the sub-blocks, obtaining a respective prediction block for the each sub-block by:
• obtaining, based on the first neighboring block, a first motion vector using a subblock-based temporal motion vector prediction mode;
• obtaining, based on the second neighboring block, a second motion vector using the subblock-based temporal motion vector prediction mode;
• and
• obtaining the respective prediction block for the sub-block based on the first motion vector and the second motion vector.