Picture prediction method and related apparatus
A picture prediction method includes: determining two pixel samples in a current picture block, and determining a candidate motion information unit set corresponding to each of the two pixel samples; determining a merged motion information unit set i including two motion information units; and predicting a pixel value of the current picture block by using an affine motion model and the merged motion information unit set i.
1. A picture processing method, comprising:
obtaining, by a picture processing apparatus, motion vectors of two pixel samples in a video frame to which a current picture block belongs; and
obtaining, by the picture processing apparatus, a motion vector of a pixel sample in the current picture block by using an affine motion model and the obtained motion vectors;
wherein the affine motion model comprises:
{
vx
=
ax
+
by
+
c
vy
=
-
bx
+
ay
+
d
,
wherein (x, y) are coordinates of the pixel sample, vx is a horizontal component of the motion vector of the pixel sample, vy is a vertical component of the motion vector of the pixel sample, and a, b, c and d are coefficients of the affine motion model.
2. The method of claim 1 , wherein when the two pixel samples comprise an upper left pixel sample of the current picture block and an upper right pixel sample of the current picture block, a is
vx
2
-
vx
0
w
,
b is
-
vy
2
-
vy
0
w
,
c is vx 0 , d is vy 0 , and the affine motion model comprises:
{
vx
=
vx
2
-
vx
0
w
x
-
vy
2
-
vy
0
w
y
+
vx
0
vy
=
vy
2
-
vy
0
w
x
+
vx
2
-
vx
0
w
y
+
vy
0
,
and wherein (vx 0 , vy 0 ) is a motion vector of the upper left pixel sample, (vx 1 , vy 1 ) is a motion vector of the upper right pixel sample, and w is a distance between the two pixel samples.
3. The method of claim 1 , wherein when the two pixel samples comprise an upper left pixel sample of the current picture block and a lower left pixel sample of the current picture block, a is
vy
2
-
vy
0
h
,
b is
vx
2
-
vx
0
h
,
c is vx 0 , d is vy 0 , and the affine motion model comprises:
{
vx
=
vy
2
-
vy
0
h
x
-
vx
2
-
vx
0
h
y
+
vx
0
vy
=
vx
2
-
vx
0
h
x
+
vy
2
-
vy
0
h
y
+
vy
0
,
and wherein (vx 0 , vy 0 ) is a motion vector of the upper left pixel sample, (vx 2 , vy 2 ) is a motion vector of the lower left pixel sample, and h is a distance between the two pixel samples.
4. The method of claim 1 , wherein after obtaining the motion vector of the pixel sample in the current picture block, the method further comprises:
performing motion compensation predictive coding on a picture block by using the motion vector of the pixel sample in the current picture block, wherein the pixel sample in the current picture block belongs to the picture block and the picture block belongs to the current picture block.
5. A picture processing apparatus comprising a processor and a non-transitory memory having instructions stored thereon, wherein the instructions, when executed by the processor, facilitate:
obtaining motion vectors of two pixel samples in a video frame to which a current picture block belongs; and
obtaining a motion vector of a pixel sample in the current picture block by using an affine motion model and the obtained motion vectors;
wherein the affine motion model comprises:
{
vx
=
ax
+
by
+
c
vy
=
-
bx
+
ay
+
d
,
wherein (x, y) are coordinates of the pixel sample, vx is a horizontal component of the motion vector of the pixel sample, vy is a vertical component of the motion vector of the pixel sample, and a, b, c and d are coefficients of the affine motion model.
6. The apparatus of claim 5 , wherein when the two pixel samples comprise an upper left pixel sample of the current picture block and an upper right pixel sample of the current picture block, a is
vx
1
-
vx
0
w
,
b is
-
vy
1
-
vy
0
w
,
c is vx 0 , d is vy 0 , and the affine motion model comprises:
{
vx
=
vx
1
-
vx
0
w
x
-
vy
1
-
vy
0
w
y
+
vx
0
vy
=
vy
1
-
vy
0
w
x
+
vx
1
-
vx
0
w
y
+
vy
0
,
and wherein (vx 0 , vy 0 ) is a motion vector of the upper left pixel sample, (vx 1 , vy 1 ) is a motion vector of the upper right pixel sample, and w is a distance between the two pixel samples.
7. The apparatus of claim 5 , wherein when the two pixel samples comprise an upper left pixel sample of the current picture block and a lower left pixel sample of the current picture block, a is
vy
2
-
vy
0
h
,
b is
vx
2
-
vx
0
h
,
c is vx 0 , d is vy 0 , and the affine motion model comprises:
{
vx
=
vy
2
-
vy
0
h
x
-
vx
2
-
vx
0
h
y
+
vx
0
vy
=
vx
2
-
vx
0
h
x
+
vy
2
-
vy
0
h
y
+
vy
0
,
and wherein (vx 0 , vy 0 ) is a motion vector of the upper left pixel sample, (vx 2 , vy 2 ) is a motion vector of the lower left pixel sample, and h is a distance between the two pixel samples.
8. The apparatus of claim 5 , wherein the instructions, when executed by the processor, further facilitate:
performing motion compensation predictive coding on a picture block by using the motion vector of the pixel sample in the current picture block, wherein the pixel sample in the current picture block belongs to the picture block and the picture block belongs to the current picture block.
9. A non-transitory computer-readable medium having processor-executable instructions stored thereon, wherein the processor-executable instructions, when executed, facilitate:
obtaining motion vectors of two pixel samples in a video frame to which a current picture block belongs; and
obtaining a motion vector of a pixel sample in the current picture block by using an affine motion model and the obtained motion vectors;
wherein the affine motion model comprises:
{
vx
=
ax
+
by
+
c
vy
=
-
bx
+
ay
+
d
,
wherein (x, y) are coordinates of the pixel sample, vx is a horizontal component of the motion vector of the pixel sample, vy is a vertical component of the motion vector of the pixel sample, and a, b, c and d are coefficients of the affine motion model.
10. The non-transitory computer-readable medium of claim 9 , wherein when the two pixel samples comprise an upper left pixel sample of the current picture block and an upper right pixel sample of the current picture block, a is
vx
1
-
vx
0
w
,
b is
-
vy
1
-
vy
0
w
,
c is vx 0 , d is vy 0 , and the affine motion model comprises:
{
vx
=
vx
1
-
vx
0
w
x
-
vy
1
-
vy
0
w
y
+
vx
0
vy
=
vy
1
-
vy
0
w
x
+
vx
1
-
vx
0
w
y
+
vy
0
,
and wherein (vx 0 , vy 0 ) is a motion vector of the upper left pixel sample, (vx 1 , vy 1 ) is a motion vector of the upper right pixel sample, and w is a distance between the two pixel samples.
11. The non-transitory computer-readable medium of claim 9 , wherein when the two pixel samples comprise an upper left pixel sample of the current picture block and a lower left pixel sample of the current picture block, a is
vy
2
-
vy
0
h
,
b is
vx
2
-
vx
0
h
,
c is vx 0 , d is vy 0 , and the affine motion model comprises:
{
vx
=
vy
2
-
vy
0
h
x
-
vx
2
-
vx
0
h
y
+
vx
0
vy
=
vx
2
-
vx
0
h
x
+
vy
2
-
vy
0
h
y
+
vy
0
,
and wherein (vx 0 , vy 0 ) is a motion vector of the upper left pixel sample, (vx 2 , vy 2 ) is a motion vector of the lower left pixel sample, and h is a distance between the two pixel samples.
12. The non-transitory computer-readable medium of claim 9 , wherein the processor-executable instructions, when executed, further facilitate:
performing motion compensation predictive coding on a picture block by using the motion vector of the pixel sample in the current picture block, wherein the pixel sample in the current picture block belongs to the picture block and the picture block belongs to the current picture block.