Planar prediction mode for visual media encoding and decoding
Devices, systems and methods for deriving intra prediction samples when the intra prediction mode for a coding block is the planar mode are described. In an aspect, a method for video coding includes selecting a first set of reference samples that are reconstructed neighboring samples of a current block, and determining a prediction value for a prediction sample of the current block by interpolating at least one of the first set and a second set of reference samples, where a reference sample of the second set of reference samples is based on a weighted sum of a first sample and a second sample from the first set of reference samples, and where the reference sample is aligned horizontally with the first sample, aligned vertically with the second sample, and positioned on an opposite side of the prediction sample with respect to either the first or the second sample.
1. A method for video coding, comprising:
selecting a first set of reference samples comprising a top reference sample and a left reference sample that are reconstructed neighboring samples of a current block;
determining a second set of reference samples based on the first set of reference samples, the second set of reference samples comprising a bottom neighboring sample and a right neighboring sample that are neighboring samples of the current block; and
determining a first prediction value for the right neighboring sample, p[W][y], of the current block based on a weighted sum from the first set of reference samples of a first sample in a first predetermined location, p[W][−1], of a same column as the right neighboring sample and a second sample in a variable location, [−1][y], in a same row as the right neighboring sample, and wherein the right neighboring sample is expressed as:
p
[
W
]
[
y
]
=
W
W
+
y
+
1
p
[
W
]
[
-
1
]
+
y
+
1
W
+
y
+
1
p
[
-
1
]
[
y
]
wherein W is a fixed value and y is a variable value,
wherein the right neighboring sample is aligned horizontally with the first sample, and
wherein the right neighboring sample is aligned vertically with the second sample.
2. The method of claim 1 , further comprising:
repeating the determining a first prediction sample for each sample of the current block;
calculating a residual of the current block using the prediction values corresponding to the each sample of the current block; and
encoding the residual into a bitstream.
3. The method of claim 1 , further comprising:
parsing a bitstream representation of the current block to determine a residual of the current block; and
reconstructing a sample of the current block based on a sum of the prediction sample and the residual.
4. The method of claim 1 , wherein a second prediction value for the bottom neighboring sample, p[x][H], of the current block is based on a weighted sum from a second predetermined location, p[−1][H] and a second variable location, p[x][−1], and wherein the bottom neighboring sample is expressed as:
p
[
x
]
[
H
]
=
H
H
+
x
+
1
p
[
-
1
]
[
H
]
+
x
+
1
H
+
x
+
1
p
[
x
]
[
-
1
]
.
5. The method of claim 1 , wherein the weighted sum comprises a first weight that multiplies the first sample and a second weight that multiplies the second sample.
6. The method of claim 5 , wherein the first and second weights are based on a distance between the right neighboring sample and the first sample and a distance between the right reference sample and the second sample.
7. The method of claim 5 , wherein the first weight is equal to the second weight.
8. An apparatus for video processing, comprising:
a processor configured to:
select a first set of reference samples comprising a top reference sample and a left reference sample that are reconstructed neighboring samples of a current block;
determine a second set of reference samples based on the first set of reference samples, the second set of reference samples comprising a bottom neighboring sample and a right neighboring sample that are neighboring samples of the current block; and
determine a first prediction value the right neighboring sample, p[W][y], of the current block based on a weighted sum from the first set of reference samples of a first sample in a first predetermined location, p[W][−1], of a same column as the right neighboring sample and a second sample in a variable location, p[−1][y], in a same row as the right neighboring sample, wherein the right neighboring sample is expressed as:
p
[
W
]
[
y
]
=
W
W
+
y
+
1
p
[
W
]
[
-
1
]
+
y
+
1
W
+
y
+
1
p
[
-
1
]
[
y
]
wherein W is a fixed value and y is a variable value,
wherein the right neighboring sample is aligned horizontally with the first sample, and
wherein the right neighboring sample is aligned vertically with the second sample.
9. The apparatus of claim 8 , wherein the processor is further configured to:
repeat the determine the first prediction sample for each sample of the current block;
calculate a residual of the current block using the prediction values corresponding to the each sample of the current block; and
encode the residual into a bitstream.
10. The apparatus of claim 8 , wherein the processor is further configured to:
parse a bitstream representation of the current block to determine a residual of the current block; and
reconstruct a sample of the current block based on a sum of the prediction sample and the residual.
11. The apparatus of claim 8 , wherein a second prediction value for the bottom neighboring sample, p[x][H], of the current block is based on a weighted sum from a second predetermined location, p[−1][H] and a second variable location, p[x][−1], and wherein the bottom neighboring sample is expressed as:
p
[
x
]
[
H
]
=
H
H
+
x
+
1
p
[
-
1
]
[
H
]
+
x
+
1
H
+
x
+
1
p
[
x
]
[
-
1
]
.