Motion vector obtaining method and apparatus
This application provides a motion vector obtaining method and apparatus. The method includes: determining a target offset vector of a block and identifier information of a target picture, wherein the block comprises at least one sub-block; determining a location of the sub-block; determining, as a target location coordinate value of a collocated sub-block, a location coordinate value obtained by performing a clipping operation on an initial location coordinate value in a range, wherein the initial location coordinate value is based on the location of the sub-block and the target offset vector; and obtaining a motion vector of the sub-block based on a motion vector corresponding to the target location coordinate value. Thus, a range of the target offset vector is limited, so that a quantity of memory read times can be reduced in a process of obtaining the motion vector of the collocated sub-block.
1. A method for inter prediction of video data, comprising:
determining a target offset vector of a block and identifier information of a target picture, wherein the block comprises a sub-block;
determining a location of the sub-block;
determining a target location coordinate value of a collocated sub-block by performing a clipping operation on an initial location coordinate value of the collocated sub-block to be within a range, wherein the initial location coordinate value is determined based on the location of the sub-block and the target offset vector, and the collocated sub-block is in the target picture and corresponds to the sub-block, wherein the target picture is indicated by the identifier information;
obtaining a motion vector of the sub-block based on a motion vector corresponding to the target location coordinate value; and
generating predicted sample values for the sub-block using the motion vector of the sub-block,
wherein the range is represented by a horizontal coordinate value range and a vertical coordinate value range, the horizontal coordinate value range comprises a minimum horizontal coordinate value and a maximum horizontal coordinate value, and the vertical coordinate value range comprises a minimum vertical coordinate value and a maximum vertical coordinate value,
wherein the determining the target location coordinate value of the collocated sub-block by performing the clipping operation on the initial location coordinate value of the collocated sub-block to be within the range, comprises:
determining the target location coordinate value according to the following formulas:
x (i,j) =Clip3(HorMin,HorMax, x (i,j) ); and
y (i,j) =Clip3(VerMin,VerMax, y (i,j) ), wherein
HorMin is the minimum horizontal coordinate value, HorMax is the maximum horizontal coordinate value, VerMin is the minimum vertical coordinate value, VerMax is the maximum vertical coordinate value, a Clip3 function is a clipping function, and the Clip3 function is defined as follows:
Clip
3
(
x
,
y
,
z
)
=
{
x
;
z
<
x
y
;
z
>
y
z
;
otherwise
.
2. The method according to claim 1 , wherein the initial location coordinate value comprises at least one of a horizontal coordinate or a vertical coordinate of the collocated sub-block;
the horizontal coordinate of the collocated sub-block is a sum of a horizontal coordinate of the location of the sub-block and a horizontal component (x off ) of the target offset vector; and
the vertical coordinate of the collocated sub-block is a sum of a vertical coordinate of the location of the sub-block and a vertical component (y off ) of the target offset vector.
3. The method according to claim 2 , wherein the initial location coordinate value of the collocated sub-block is derived as follows:
{
x
(
i
,
j
)
=
x
+
M
×
i
+
M
2
+
x
off
y
(
i
,
j
)
=
y
+
N
×
j
+
N
2
+
y
off
,
wherein
(x, y) represents location coordinates of the block, i represents the sub-block is an i th sub-block in the block in a horizontal direction, j represents the sub-block is a j th sub-block in the block in a vertical direction, (x off , y off ) represents values of components of the target offset vector in the horizontal direction and the vertical direction, respectively, M and N are respectively a width and a height of the sub-block, and (x (i,j) , y (i,j) ) represents the initial location coordinate value of the collocated sub-block that corresponds to the (i, j) th sub-block.
4. The method according to claim 3 , wherein (x, y) represents location coordinates of a top-left vertex of the block, and (x (i,j) , y (i,j) ) represents location coordinates of a center point or a top-left vertex of the collocated sub-block that corresponds to the (i, j) th sub-block.
5. The method according to claim 1 , wherein the sub-block is an (i, j) th sub-block in the block, location coordinates of the (i, j) th sub-block are represented by
(
x
+
M
×
i
+
M
2
,
y
+
N
×
j
+
N
2
)
(x, y) represents location coordinates of the block, i represents the sub-block is the i th sub-block in the block in a horizontal direction, j represents the sub-block is the j th sub-block in the block in a vertical direction, and M and N are respectively a width and a height of the sub-block.
6. The method according to claim 1 , wherein the range is determined based on a location and a size of a coding tree unit (CTU) in which the block is included, a size of the target picture, and an extension range of a temporal motion vector field of the CTU.
7. The method according to claim 6 , wherein the range is determined according to the following formulas:
HorMin=Max( CTUX −Range L, 0);
HorMax=Min( CTUX+CTUW +Range R −1, Pic W −1);
VerMin=Max( CTUY −Range U, 0); and
VerMax=Min( CTUY+CTUH +Range B −1, Pic H −1), wherein
HorMin and HorMax are respectively a minimum value and a maximum value of the range in a horizontal direction, VerMin and VerMax are respectively a minimum value and a maximum value of the range in a vertical direction, CTUX and CTUY represent location coordinates of the CTU in which the block is included, CTUW and CTUH respectively represent a width and a height of the CTU in which the block is included, RangeU, RangeB, RangeL, and RangeR respectively represent extension range values that are in an upward direction, a downward direction, a left direction, and a right direction and that are of the temporal motion vector field of the CTU in which the block is included, and PicW and PicH are respectively a width and a height of the target picture.
8. The method according to claim 7 , wherein a value of the extension range of the temporal motion vector field of the CTU (RangeU, RangeB, RangeL, RangeR) is a predetermined value.
9. The method according to claim 1 , wherein the obtaining a motion vector of the sub-block based on a motion vector corresponding to the target location coordinate value comprises:
scaling the motion vector corresponding to the target location coordinate value based on a picture order count (POC) of a picture comprising the sub-block, a POC of a target reference picture of the picture comprising the sub-block, a POC of the target picture, and a POC of a target reference picture of the target picture, to obtain the motion vector of the sub-block.
10. The method according to claim 9 , wherein the motion vector of the sub-block is obtained according to the following formula:
MV
s
=
P
1
-
P
2
P
3
-
P
4
×
MV
,
wherein
P 1 represents the POC of the picture comprising the sub-block, P 2 represents the POC of the target reference picture of the picture comprising the sub-block, P 3 represents the POC of the target picture, P 4 represents the POC of the target reference picture of the target picture, MV represents the motion vector corresponding to the target location coordinate value, and MV s represents the motion vector of the sub-block.
11. The method according to claim 1 , wherein the determining the target offset vector of the block comprises:
determining a motion vector of a spatially neighboring block of the block as the target offset vector.
12. An apparatus, comprising:
a non-transitory computer-readable medium configured to store computer-readable instructions; and
a processor in communication with the non-transitory computer-readable medium and configured to execute the computer-readable instructions stored in the non-transitory computer-readable medium thereby performing operations comprising:
determining a target offset vector of a block and identifier information of a target picture, wherein the block comprises a sub-block;
determining a location of the sub-block;
determining a target location coordinate value of a collocated sub-block by performing a clipping operation on an initial location coordinate value of the collocated sub-block to be within a range, wherein the initial location coordinate value is determined based on the location of the sub-block and the target offset vector, and the collocated sub-block is in the target picture and corresponds to the sub-block, wherein the target picture is indicated by the identifier information;
obtaining a motion vector of the sub-block based on a motion vector corresponding to the target location coordinate value;
generating predicted sample values for the sub-block using the motion vector of the sub-block,
wherein the range is represented by a horizontal coordinate value range and a vertical coordinate value range, the horizontal coordinate value range comprises a minimum horizontal coordinate value and a maximum horizontal coordinate value, and the vertical coordinate value range comprises a minimum vertical coordinate value and a maximum vertical coordinate value; and
determining the target location coordinate value according to the following formulas:
x (i,j) =Clip3(HorMin,HorMax, x (i,j) ; and
y (i,j) =Clip3(VerMin,VerMax, y (i,j) ), wherein
HorMin is the minimum horizontal coordinate value, HorMax is the maximum horizontal coordinate value, VerMin is the minimum vertical coordinate value, VerMax is the maximum vertical coordinate value, a Clip3 function is a clipping function, and the Clip3 function is defined as follows:
Clip
3
(
x
,
y
,
z
)
=
{
x
;
z
<
x
y
;
z
>
y
z
;
otherwise
.
13. The apparatus according to claim 12 , wherein the initial location coordinate value comprises at least one of a horizontal coordinate or a vertical coordinate of the collocated sub-block;
a horizontal coordinate of the collocated sub-block is a sum of a horizontal coordinate of the location of the sub-block and a horizontal component (x off ) of the target offset vector; and
a vertical coordinate of the collocated sub-block is a sum of a vertical coordinate of the location of the sub-block and a vertical component (y off ) of the target offset motion vector.
14. The apparatus according to claim 13 , wherein the initial location coordinate value of the collocated sub-block is derived as follows:
{
x
(
i
,
j
)
=
x
+
M
×
i
+
M
2
+
x
off
y
(
i
,
j
)
=
y
+
N
×
j
+
N
2
+
y
off
,
wherein
(x, y) represents location coordinates of the block, i represents the sub-block is an i th sub-block in the block in a horizontal direction, j represents the sub-block is a j th sub-block in the block in a vertical direction, (x off , y off ) represents values of components of the target offset motion vector in the horizontal direction and the vertical direction, respectively, M and N are respectively a width and a height of the sub-block, and (x (i,j) , y (i,j) ) represents the initial location coordinate value of the collocated sub-block that corresponds to the (i, j) th sub-block.
15. The apparatus according to claim 14 , wherein (x, y) represents location coordinates of a top-left vertex of the block, and (x (i,j) , y (i,j) ) represents location coordinates of a center point or a top-left vertex of the collocated sub-block that corresponds to the (i, j) th sub-block.
16. The apparatus according to claim 12 , wherein the sub-block is an (i, j) th sub-block in the block, location coordinates of the (i, j) th sub-block are represented by
(
x
+
M
×
i
+
M
2
,
y
+
N
×
j
+
N
2
)
(x, y) represents location coordinates of the block, i represents the sub-block is the i th sub-block in the block in a horizontal direction, j represents the sub-block is the j th sub-block in the block in a vertical direction, and M and N are respectively a width and a height of the sub-block.
17. The apparatus according to claim 12 , wherein the range is determined based on a location and a size of a coding tree unit (CTU) in which the block is included, a size of the target picture, and an extension range of a temporal motion vector field of the CTU.
18. The apparatus according to claim 17 , wherein the range is determined according to the following formulas:
HorMin=Max( CTUX −Range L, 0);
HorMax=Min( CTUX+CTUW +Range R− 1, Pic W −1);
VerMin=Max( CTUY −Range U, 0); and
VerMax=Min( CTUY+CTUH +Range B −1, Pic H −1), wherein
HorMin and HorMax are respectively a minimum value and a maximum value of the range in a horizontal direction, VerMin and VerMax are respectively a minimum value and a maximum value of the range in a vertical direction, CTUX and CTUY represent location coordinates of the CTU in which the block is included, CTUW and CTUH represent a width and a height of the CTU in which the block is included, RangeU, RangeB, RangeL, and RangeR respectively represent extension range values that are in an upward direction, a downward direction, a left direction, and a right direction and that are of the temporal motion vector field of the CTU in which the block is included, and PicW and PicH are respectively a width and a height of the target picture.
19. The apparatus according to claim 18 , wherein a value of the extension range of the temporal motion vector field of the CTU (RangeU, RangeB, RangeL, RangeR) is a predetermined value.
20. The apparatus according to claim 12 , wherein the operations further comprise:
scaling the motion vector corresponding to the target location coordinate value based on a picture order count (POC) of a picture comprising the sub-block, a POC of a target reference picture of the picture comprising the sub-block, a POC of the target picture, and a POC of a target reference picture of the target picture, to obtain the motion vector of the sub-block.
21. The apparatus according to claim 20 , wherein the motion vector of the sub-block is obtained according to the following formula:
MV
s
=
P
1
-
P
2
P
3
-
P
4
×
MV
,
wherein
P 1 represents the POC of the picture comprising the sub-block, P 2 represents the POC of the target reference picture of the picture comprising the sub-block, P 3 represents the POC of the target picture, P 4 represents the POC of the target reference picture of the target picture, MV represents the motion vector corresponding to the target location coordinate value, and MV s represents the motion vector of the sub-block.
22. The apparatus according to claim 12 , wherein the operations further comprise:
determining a motion vector of a spatially neighboring block of the block as the target offset vector.
23. The apparatus according to claim 12 , wherein the apparatus is a decoding apparatus or an encoding apparatus.
24. A non-transitory computer-readable medium, comprising program code, which, when executed by a processor, causes the processor to perform operations comprising:
determining a target offset vector of a block and identifier information of a target picture, wherein the block comprises a sub-block;
determining a location of the sub-block;
determining a target location coordinate value of a collocated sub-block by performing a clipping operation on an initial location coordinate value of the collocated sub-block to be within a range, wherein the initial location coordinate value is determined based on the location of the sub-block and the target offset vector, and the collocated sub-block is in the target picture and corresponds to the sub-block, wherein the target picture is indicated by the identifier information;
obtaining a motion vector of the sub-block based on a motion vector corresponding to the target location coordinate value; and
generating predicted sample values for the sub-block using the motion vector of the sub-block,
wherein the range is represented by a horizontal coordinate value range and a vertical coordinate value range, the horizontal coordinate value range comprises a minimum horizontal coordinate value and a maximum horizontal coordinate value, and the vertical coordinate value range comprises a minimum vertical coordinate value and a maximum vertical coordinate value,
wherein the determining the target location coordinate value of a collocated sub-block by performing the clipping operation on the initial location coordinate value of the collocated sub-block to be within the range, comprises:
determining the target location coordinate value according to the following formulas:
x (i,j) =Clip3(HorMin,HorMax, x (i,j) ); and
y (i,j) =Clip3(VerMin,VerMax, y (i,j) ), wherein
HorMin is the minimum horizontal coordinate value, HorMax is the maximum horizontal coordinate value, VerMin is the minimum vertical coordinate value, VerMax is the maximum vertical coordinate value, a Clip3 function is a clipping function, and the Clip3 function is defined as follows:
C
l
i
p
3
(
x
,
y
,
z
)
=
{
x
;
z
<
x
y
;
z
>
y
z
;
otherwise
.