Encoder, decoder and corresponding methods for adaptive loop filtering

Patent metadata

Field	Value
Publication number	US-12267491-B2
Application number	US-202217871753-A
Country	US
Kind code	B2
Filing date	Jul 22, 2022
Priority date	Jan 24, 2020
Publication date	Apr 1, 2025
Grant date	Apr 1, 2025

Abstract

A method of filtering a sample value of a picture and a video decoding apparatus are disclosed, the method comprises: obtaining a reconstructed sample value for a block of the picture; obtaining filter coefficients for adaptive loop filtering according to a bitstream; obtaining a sum value according to the filter coefficients and the reconstructed sample value for the block; rounding the sum value according to a vertical position of a sample in the block, to obtain a rounded sum value; obtaining a filter reconstructed sample value for the block according to the rounded sum value.

First claim

The invention claimed is:

1. A method of filtering a sample value of a picture, wherein the method comprises:
obtaining a reconstructed sample value for a block of the picture;
obtaining filter coefficients for adaptive loop filtering according to a bitstream;
obtaining a sum value according to the filter coefficients, and the reconstructed sample value for the block;
rounding the sum value according to a vertical position of a sample in the block to obtain a rounded sum value, wherein the vertical position of the sample in the block is a luma sample position, and wherein the rounded sum value equals curr+((sum+roundY)>>alfShiftY) and satisfies the following table:

Condition
roundY
alfShiftY
y1
y2
y3

(y = = CtbSizeY − 5 | | y = =
512
10
0
0
0

CtbSizeY − 4) &&

(applyAlfLineBufBoundary = = 1)

(y = = CtbSizeY − 6 | | y = =
 64
 7
1
1
1

CtbSizeY − 3) &&

(applyAlfLineBufBoundary = = 1)

(y = = CtbSizeY − 7 | | y = =
 64
 7
1
2
2

CtbSizeY − 2) &&

(applyAlfLineBufBoundary = = 1)

Otherwise
 64
 7
1
2
3

wherein
y represents the vertical position of the sample;
curr represents the reconstructed sample value for the block;
sum represents the sum value;
CtbSizeY represents a size of a coding tree block (CTB) corresponding to the block;
v 1 , y 2 and y 3 represent vertical sample position offsets indicating other reconstructed sample values involved in the obtaining the sum value;
roundY and alfShiftY are variables that are determined based on the vertical position of the sample;
apply AlfLineBufBoundary is an index and derived as follows:
if a bottom boundary of the coding tree block is a bottom boundary of the picture and pic_height_in_luma_samples−yCtb≤CtbSizeY−4,

apply AlfLineBufBoundary is set equal to 0, otherwise,
apply AlfLineBufBoundary is set equal to 1,
wherein the pic_height_in_luma_samples represents a height of the picture in luma samples, and (xCtb, yCtb) represents a location of the CTB; and

obtaining a filter reconstructed sample value for the block according to the rounded sum value.

2. The method of claim 1 , wherein roundY is equal to 512 and alfShiftY is equal to 10 when the vertical position of the sample satisfies any one of the below conditions:
y==CtbSizeY−t-1, or
y==CtbSizeY−t, or
y==ctbHeightC−t, or
y==ctbHeightC−t-1,
wherein y is the vertical position of the sample, CtbSizeY is a coding tree block (CTB) size, ctbHeightC is a height of the CTB, and t is an integer value, wherein t is 2, 3, 4, 5 or 6.

3. The method of claim 1 , wherein roundY is equal to 64 and alfShiftY is equal to 7 when the vertical position of the sample does not satisfy any one of the below conditions:
y==CtbSizeY−t-1, or
y==CtbSizeY−t, or
y==ctbHeightC−t, or
y==ctbHeightC−t-1,
wherein y is the vertical position of the sample, CtbSizeY is a coding tree block (CTB) size, ctbHeightC is a height of the CTB, and t is an integer value, wherein t is 2, 3, 4, 5 or 6.

4. The method of claim 1 , wherein when the vertical position of the sample in the block is equal to 0, alfShiftY is equal to 10.

5. The method of claim 1 , wherein when the vertical position of the sample in the block is not equal to 0, alfShiftY is equal to 7.

6. The method of claim 1 , wherein the rounding the sum value comprises:
adding the value roundY which is determined based on the vertical position of the sample to the sum value in order to obtain an added sum value, and
shifting the added sum value based on the vertical position of the sample.

7. A decoder, comprising:
one or more processors; and
a non-transitory computer-readable storage medium coupled to the one or more processors and storing programming for execution by the one or more processors, wherein the programming, when executed by the processors, causes the decoder to perform operations comprising:
obtaining a reconstructed sample value for a block of a picture;
obtaining filter coefficients for adaptive loop filtering according to a bitstream;
obtaining a sum value according to the filter coefficients and the reconstructed sample value for the block;
rounding the sum value according to a vertical position of a sample in the block, to obtain a rounded sum value, wherein the vertical position of the sample in the block is a luma sample position, and wherein the rounded sum value equals curr+((sum+roundY)>>alfShiftY) and satisfies the following table:

Condition
roundY
alfShiftY
y1
y2
y3

(y = = CtbSizeY − 5 | | y = =
512
10
0
0
0

CtbSizeY − 4) &&

(applyAlfLineBufBoundary = = 1)

(y = = CtbSizeY − 6 | | y = =
 64
 7
1
1
1

CtbSizeY − 3) &&

Assignees

• Huawei Tech Co Ltd

Inventors

• Kotra Anand Meher
• Esenlik Semih
• Gao Han
• Wang Biao
• Alshina Elena Alexandrovna

Classifications

• H04N19/176 Primary

  the region being a block, e.g. a macroblock · CPC title

• H04N19/96

  Tree coding, e.g. quad-tree coding · CPC title

• H04N19/167

  Position within a video image, e.g. region of interest [ROI] · CPC title

• H04N19/86

  involving reduction of coding artifacts, e.g. of blockiness · CPC title

• H04N19/132

  Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking · CPC title