Image signal encoding/decoding method and non-transitory computer-readable medium
View Patent ↗An image decoding method according to the present disclosure comprises the steps of: generating a merge candidate list in a current block; specifying one of a plurality of merge candidates included in the merge candidate list; deriving a first affine seed vector and a second affine seed vector of the current block on the basis of a first affine seed vector and a second affine seed vector of the specified merge candidate; deriving an affine vector for a subblock in the current block, using the first affine seed vector and the second affine seed vector of the current block; and performing motion compensation prediction for the subblock on the basis of the affine vector.
1. A method for video decoding, comprising:
generating a merge candidate list for a current block;
specifying one among a plurality of merge candidates included in the merge candidate list;
deriving a first affine seed vector and a second affine seed vector of the current block based on a first affine seed vector and a second affine seed vector of the specified merge candidate; and
deriving an affine vector for a subblock in the current block by using the first affine seed vector and the second affine seed vector of the current block, wherein the subblock is a region of a size smaller than that of the current block;
wherein the first affine seed vector and the second affine seed vector of the merge candidate are derived based on motion information of a neighboring block adjacent to the current block; and
in response to the neighboring block being included in a coding tree unit different from a coding tree unit of the current block, the first affine seed vector and the second affine seed vector of the merge candidate are derived based on motion vectors of a bottom-left subblock and a bottom-right subblock of the neighboring block.
2. The method according to claim 1 , wherein the bottom-left subblock includes a bottom-left reference sample positioned at a bottom-left corner of the neighboring block, and the bottom-right subblock includes a bottom-right reference sample located at a bottom-left corner of the neighboring block.
3. The method according to claim 2 , wherein the first affine seed vector and the second affine seed vector of the merge candidate are derived based on a value obtained based on a shift operation performed on a value of difference between motion vectors of the bottom-left subblock and the bottom-right subblock using a scaling factor, and the scaling factor is derived based on a value obtained by adding a horizontal distance between the bottom-left reference sample and the bottom-right reference sample and an offset.
4. The method according to claim 1 , wherein the first affine seed vector and the second affine seed vector of the current block are derived using a third affine seed vector for a bottom-left control point and a fourth affine seed vector for a bottom-right control point of a neighboring affine vector, wherein
v
0
x
=
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
x
0
-
x
n
2
)
-
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
y
0
-
y
n
2
)
+
n
v
2
x
v
0
y
=
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
x
0
-
x
n
2
)
-
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
y
0
-
y
n
2
)
+
n
v
2
y
v
1
x
=
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
x
1
-
x
n
2
)
-
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
y
1
-
y
n
2
)
+
n
v
2
x
v
1
y
=
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
x
1
-
x
n
2
)
-
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
y
1
-
y
n
2
)
+
n
v
2
y
where (x n2 , y n2 ) denotes coordinates of a bottom-left control point of an affine neighboring block, and (x n3 , y n3 ) denotes coordinates of a bottom-right control point of the affine neighboring block;
(x 0 , y 0 ) denotes coordinates of a top-left control point of the current block, and (x 1 , y 1 ) denotes coordinates of a top-right control point of the current block;
(nv 2x , nv 2y ) denotes an affine seed vector of a bottom-left control point of the third affine seed vector, and (nv 3x , nv 3y ) denotes an affine seed vector of a bottom-right control point of the fourth affine seed vector; and
(v 0x , v 0y ) denotes an affine seed vector of a top-left control point of the first affine seed vector, and (v 1x , v 1y ) denotes an affine seed vector of a top-right control point of the second affine seed vector.
5. The method according to claim 1 , wherein a top boundary of the current block is in contact with a top boundary of the coding tree unit of the current block.
6. A method for video encoding, comprising:
generating a merge candidate list for a current block;
specifying one among a plurality of merge candidates included in the merge candidate list;
deriving a first affine seed vector and a second affine seed vector of the current block based on a first affine seed vector and a second affine seed vector of the specified merge candidate; and
deriving an affine vector for a subblock in the current block by using the first affine seed vector and the second affine seed vector of the current block, wherein the subblock is a region of a size smaller than that of the current block;
wherein the first affine seed vector and the second affine seed vector of the merge candidate are derived based on motion information of a neighboring block adjacent to the current block; and
in response to the neighboring block being included in a coding tree unit different from a coding tree unit of the current block, the first affine seed vector and the second affine seed vector of the merge candidate are derived based on motion vectors of a bottom-left subblock and a bottom-right subblock of the neighboring block.
7. The method according to claim 6 , wherein the bottom-left subblock includes a bottom-left reference sample positioned at a bottom-left corner of the neighboring block, and the bottom-right subblock includes a bottom-right reference sample located at a bottom-left corner of the neighboring block.
8. The method according to claim 7 , wherein the first affine seed vector and the second affine seed vector of the merge candidate are derived based on a value obtained based on a shift operation performed on a value of difference between motion vectors of the bottom-left subblock and the bottom-right subblock using a scaling factor, and the scaling factor is derived based on a value obtained by adding a horizontal distance between the bottom-left reference sample and the bottom-right reference sample and an offset.
9. The method according to claim 6 , wherein the first affine seed vector and the second affine seed vector of the current block are derived using a third affine seed vector for a bottom-left control point and a fourth affine seed vector for a bottom-right control point of a neighboring affine vector, wherein
v
0
x
=
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
x
0
-
x
n
2
)
-
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
y
0
-
y
n
2
)
+
n
v
2
x
v
0
y
=
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
x
0
-
x
n
2
)
-
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
y
0
-
y
n
2
)
+
n
v
2
y
v
1
x
=
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
x
1
-
x
n
2
)
-
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
y
1
-
y
n
2
)
+
n
v
2
x
v
1
y
=
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
x
1
-
x
n
2
)
-
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
y
1
-
y
n
2
)
+
n
v
2
y
where (x n2 , y n2 ) denotes coordinates of a bottom-left control point of an affine neighboring block, and (x n3 , y n3 ) denotes coordinates of a bottom-right control point of the affine neighboring block;
(x 0 , y 0 ) denotes coordinates of a top-left control point of the current block, and (x 1 , y 1 ) denotes coordinates of a top-right control point of the current block;
(nv 2x , nv 2y ) denotes an affine seed vector of a bottom-left control point of the third affine seed vector, and (nv 3x , nv 3y ) denotes an affine seed vector of a bottom-right control point of the fourth affine seed vector; and
(v 0x , v 0y ) denotes an affine seed vector of a top-left control point of the first affine seed vector, and (v 1x , v 1y ) denotes an affine seed vector of a top-right control point of the second affine seed vector.
10. The method according to claim 6 , wherein the neighboring block is adjacent to the top boundary of the current block.
11. A non-transitory computer-readable medium having stored thereon instructions that when executed by a processor, cause the processor to perform a video decoding method, the method comprising:
generating a merge candidate list for a current block;
specifying one among a plurality of merge candidates included in the merge candidate list;
deriving a first affine seed vector and a second affine seed vector of the current block based on a first affine seed vector and a second affine seed vector of the specified merge candidate; and
deriving an affine vector for a subblock in the current block by using the first affine seed vector and the second affine seed vector of the current block, wherein the subblock is a region of a size smaller than that of the current block;
wherein the first affine seed vector and the second affine seed vector of the merge candidate are derived based on motion information of a neighboring block adjacent to the current block; and
in response to the neighboring block being included in a coding tree unit different from a coding tree unit of the current block, the first affine seed vector and the second affine seed vector of the merge candidate are derived based on motion vectors of a bottom-left subblock and a bottom-right subblock of the neighboring block.
12. The non-transitory computer-readable medium according to claim 11 , wherein the bottom-left subblock includes a bottom-left reference sample positioned at a bottom-left corner of the neighboring block, and the bottom-right subblock includes a bottom-right reference sample located at a bottom-left corner of the neighboring block.
13. The non-transitory computer-readable medium according to claim 12 , wherein the first affine seed vector and the second affine seed vector of the merge candidate are derived based on a value obtained based on a shift operation performed on a value of difference between motion vectors of the bottom-left subblock and the bottom-right subblock using a scaling factor, and the scaling factor is derived based on a value obtained by adding a horizontal distance between the bottom-left reference sample and the bottom-right reference sample and an offset.
14. The non-transitory computer-readable medium according to claim 11 , wherein the first affine seed vector and the second affine seed vector of the current block are derived using a third affine seed vector for a bottom-left control point and a fourth affine seed vector for a bottom-right control point of a neighboring affine vector, wherein
v
0
x
=
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
x
0
-
x
n
2
)
-
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
y
0
-
y
n
2
)
+
n
v
2
x
v
0
y
=
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
x
0
-
x
n
2
)
-
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
y
0
-
y
n
2
)
+
n
v
2
y
v
1
x
=
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
x
1
-
x
n
2
)
-
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
y
1
-
y
n
2
)
+
n
v
2
x
v
1
y
=
(
n
v
3
y
-
n
v
2
y
)
(
x
n
3
-
x
n
2
)
(
x
1
-
x
n
2
)
-
(
n
v
3
x
-
n
v
2
x
)
(
x
n
3
-
x
n
2
)
(
y
1
-
y
n
2
)
+
n
v
2
y
where (x n2 , y n2 ) denotes coordinates of a bottom-left control point of an affine neighboring block, and (x n3 , y n3 ) denotes coordinates of a bottom-right control point of the affine neighboring block;
(x 0 , y 0 ) denotes coordinates of a top-left control point of the current block, and (x 1 , y 1 ) denotes coordinates of a top-right control point of the current block;
(nv 2x , nv 2y ) denotes an affine seed vector of a bottom-left control point of the third affine seed vector, and (nv 3x , nv 3y ) denotes an affine seed vector of a bottom-right control point of the fourth affine seed vector; and
(v 0x , v 0y ) denotes an affine seed vector of a top-left control point of the first affine seed vector, and (v 1x , v 1y ) denotes an affine seed vector of a top-right control point of the second affine seed vector.
15. The non-transitory computer-readable medium according to claim 11 , wherein a top boundary of the current block is in contact with a top boundary of the coding tree unit of the current block.