Intra block copy vector difference wrap around
A decoder receives, from a bitstream, an indication of a block vector predictor (BVP) and a first block vector difference (BVD) for a current block. The decoder determines a second BVD based on: the first BVD and whether a location displaced from the current block by a sum of the BVP and the first BVD is outside a reference region. The decoder calculates a block vector (BV) based on a sum of the second BVD and the BVP. The decoder reconstructs the current block based on a reference block indicated by the BV.
1. A method comprising:
receiving, from a bitstream, an indication of a block vector predictor (BVP) and a first block vector difference (BVD) for a current block;
based on a location, displaced from the current block by a sum of the BVP and the first BVD, being outside a reference region:
selecting one component of the first BVD to adjust;
determining a first component, of a second BVD and corresponding to the selected component of the first BVD, to be equal to the adjusted component of the first BVD; and
determining a second component of the second BVD to be equal to a non-selected component of the first BVD;
calculating a block vector (BV) based on a sum of the second BVD and the BVP; and
reconstructing the current block based on a reference block indicated by the BV.
2. The method of claim 1 , wherein the reference region comprises a plurality of rows and a plurality of columns, and wherein:
at least one row of the plurality of rows has a different length than that of another row of the plurality of rows; or
at least one column of the plurality of columns has a different length than that of another column of the plurality of columns.
3. The method of claim 1 , wherein the first component of the second BVD is determined as having:
a magnitude based on a difference between a length of the reference region and a magnitude of the selected component of the first BVD; and
a sign opposite of that of the selected component of the first BVD.
4. The method of claim 3 , wherein the length of the reference region is along a direction of the selected component of the first BVD.
5. The method of claim 4 , wherein the direction is horizontal based on the selected component being a horizontal component and the direction is vertical based on the selected component being a vertical component.
6. The method of claim 1 , wherein the selected component of the first BVD is in a direction that results in a magnitude of the adjusted component of the first BVD being greater than half the length of the reference region.
7. The method of claim 6 , wherein the length of the reference region is along a horizontal direction based on the selected component of the first BVD being a horizontal component, and wherein the length of the reference region is along a vertical direction based on the selected component of the first BVD being a vertical component.
8. A video decoder comprising:
one or more processors; and
memory storing instructions that, when executed by the one or more processors, cause the video decoder to:
receive, from a bitstream, an indication of a block vector predictor (BVP) and a first block vector difference (BVD) for a current block;
based on a location, displaced from the current block by a sum of the BVP and the first BVD, being outside a reference region:
select one component of the first BVD to adjust;
determine a first component, of a second BVD and corresponding to the selected component of the first BVD, to be equal to the adjusted component of the first BVD; and
determine a second component of the second BVD to be equal to a non-selected component of the first BVD;
calculate a block vector (BV) based on a sum of the second BVD and the BVP; and
reconstruct the current block based on a reference block indicated by the BV.
9. The video decoder of claim 8 , wherein the reference region comprises a plurality of rows and a plurality of columns, and wherein:
at least one row of the plurality of rows has a different length than that of another row of the plurality of rows; or
at least one column of the plurality of columns has a different length than that of another column of the plurality of columns.
10. The video decoder of claim 8 , wherein the first component of the second BVD is determined as having:
a magnitude based on a difference between a length of the reference region and a magnitude of the selected component of the first BVD; and
a sign opposite of that of the selected component of the first BVD.
11. The video decoder of claim 10 , wherein the length of the reference region is along a direction of the selected component of the first BVD.
12. The video decoder of claim 11 , wherein the direction is horizontal based on the selected component being a horizontal component and the direction is vertical based on the selected component being a vertical component.
13. The video decoder of claim 8 , wherein the selected component of the first BVD is in a direction that results in a magnitude of the adjusted component of the first BVD being greater than half the length of the reference region.
14. The video decoder of claim 13 , wherein the length of the reference region is along a horizontal direction based on the selected component of the first BVD being a horizontal component, and wherein the length of the reference region is along a vertical direction based on the selected component of the first BVD being a vertical component.
15. A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors of a decoder, cause the decoder to:
receive, from a bitstream, an indication of a block vector predictor (BVP) and a first block vector difference (BVD) for a current block;
based on a location, displaced from the current block by a sum of the BVP and the first BVD, being outside a reference region:
select one component of the first BVD to adjust;
determine a first component, of a second BVD and corresponding to the selected component of the first BVD, to be equal to the adjusted component of the first BVD; and
determine a second component of the second BVD to be equal to a non-selected component of the first BVD;
calculate a block vector (BV) based on a sum of the second BVD and the BVP; and
reconstruct the current block based on a reference block indicated by the BV.
16. The non-transitory computer-readable medium of claim 15 , wherein the first component of the second BVD is determined as having with:
a magnitude based on a difference between a length of the reference region and a magnitude of the selected component of the first BVD; and
a sign opposite of that of the selected component of the first BVD.
17. The non-transitory computer-readable medium of claim 16 , wherein the length of the reference region is along a direction of the selected component of the first BVD.
18. The non-transitory computer-readable medium of claim 17 , wherein the direction is horizontal based on the selected component being a horizontal component and the direction is vertical based on the selected component being a vertical component.
19. The non-transitory computer-readable medium of claim 15 , wherein the selected component of the first BVD is in a direction that results in a magnitude of the adjusted component of the first BVD being greater than half the length of the reference region.
20. The non-transitory computer-readable medium of claim 19 , wherein the length of the reference region is along a horizontal direction based on the selected component of the first BVD being a horizontal component, and wherein the length of the reference region is along a vertical direction based on the selected component of the first BVD being a vertical component.