Method and apparatus for sub-block based temporal motion vector prediction

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

acquiring a current picture from a coded video bitstream;

identifying, for a current block included in the current picture, a reference block included in a reference picture, the reference picture being different from the current picture, the current block being divided into a first plurality of sub-blocks (CBSBs) that are coded in an inter mode, the reference block having a second plurality of sub-blocks (RBSBs), each CBSB corresponding to at least one RBSB;

for each CBSB, in response to a corresponding RBSB being coded in an intra mode or intra block copy mode, setting a first motion vector predictor of the corresponding CBSB to a default motion vector predictor, the default motion vector predictor being a zero motion vector;

for each CBSB, in response to a corresponding RBSB being coded in an inter mode, setting the first motion vector predictor of the corresponding CBSB using a second motion vector predictor of the corresponding RBSB as the first motion vector predictor of the corresponding CBSB; and

decoding each CBSB by performing sub-block based temporal motion vector prediction on each CBSB based on the first motion vector predictor for each CBSB.

2. The method of claim 1 , wherein for each CBSB, in response to the corresponding RBSB being coded in the inter mode, the first motion vector predictor is a scaled version of the second motion vector predictor of the corresponding RBSB.

3. The method of claim 1 , wherein the reference block is identified in accordance with a motion vector predictor associated with a block adjacent to the current block.

4. The method of claim 1 , wherein the reference picture is a first reference picture from a sequence of reference pictures associated with the current picture.

5. A video decoder for video decoding, comprising:

processing circuitry configured to:

acquire a current picture from a coded video bitstream,

identify, for a current block included in the current picture, a reference block included in a reference picture, the reference picture being different from the current picture, the current block being divided into a first plurality of sub-blocks (CBSBs) that are coded in an inter mode, the reference block having a second plurality of sub-blocks (RBSBs), each CBSB corresponding to at least one RBSB,

for each CBSB, in response to a corresponding RBSB being coded in an intra mode or intra block copy mode, set a first motion vector predictor of the corresponding CBSB to a default motion vector predictor, the default motion vector predictor being a zero motion vector,

for each CBSB, in response to a corresponding RBSB being coded in an inter mode, set the first motion vector predictor of the corresponding CBSB using a second motion vector predictor of the corresponding RBSB as the first motion vector predictor of the corresponding CBSB, and

decode each CBSB by performing sub-block based temporal motion vector prediction on each CBSB based on the first motion vector predictor for each CBSB.

6. The video decoder of claim 5 , wherein for each CBSB, in response to the corresponding RBSB being coded in the inter mode, the first motion vector predictor is a scaled version of the second motion vector predictor of the corresponding RBSB.

7. The video decoder of claim 5 , wherein the reference block is identified in accordance with a motion vector predictor associated with a block adjacent to the current block.

8. The video decoder of claim 5 , wherein the reference picture is a first reference picture from a sequence of reference pictures associated with the current picture.

9. A non-transitory computer readable medium having instructions stored therein, which when executed by a processor in a video decoder causes the processor to execute a method comprising:

acquiring a current picture from a coded video bitstream;

identifying, for a current block included in the current picture, a reference block included in a reference picture, the reference picture being different from the current picture, the current block being divided into a first plurality of sub-blocks (CBSBs) that are coded in an inter mode, the reference block having a second plurality of sub-blocks (RBSBs), each CBSB corresponding to at least one RBSB;

for each CBSB, in response to a corresponding RBSB being coded in an intra mode or intra block copy mode, setting a first motion vector predictor of the corresponding CBSB to a default motion vector predictor, the default motion vector predictor being a zero motion vector;

for each CBSB, in response to a corresponding RBSB being coded in an inter mode, setting the first motion vector predictor of the corresponding CBSB using a second motion vector predictor of the corresponding RBSB as the first motion vector predictor of the corresponding CBSB; and

decoding each CBSB by performing sub-block based temporal motion vector prediction on each CBSB based on the first motion vector predictor for each CBSB.

10. The non-transitory computer readable medium of claim 9 , wherein for each CBSB, in response to the corresponding RBSB being coded in the inter mode, the first motion vector predictor is a scaled version of the second motion vector predictor of the corresponding RBSB.

11. The non-transitory computer readable medium of claim 9 , wherein the reference block is identified in accordance with a motion vector predictor associated with a block adjacent to the current block.