Image or video coding based on escape binarization in palette mode

Patent metadata

Field	Value
Publication number	US-12177485-B2
Application number	US-202017639129-A
Country	US
Kind code	B2
Filing date	Aug 31, 2020
Priority date	Aug 29, 2019
Publication date	Dec 24, 2024
Grant date	Dec 24, 2024

Abstract

According to the disclosure of the present document, an escape value may be effectively binarized on the basis of a maximum value of an escape value, which may be inferred by a quantization parameter in a palette mode. Accordingly, in a palette mode, the length of a bin may be adjusted or the number bits may be saved, according to a maximum value of an escape value, whereby coding efficiency may be improved.

First claim

What is claimed is:

1. An image decoding method performed by a decoding apparatus, the image decoding method comprising:
obtaining image information including information on a quantized escape value in a palette mode and information on a quantization parameter from a bitstream;
deriving an escape value for a current block based on the information on the quantized escape value and the information on the quantization parameter; and
generating a reconstructed sample of the current block based on the escape value,
wherein the information on the quantized escape value is obtained based on a binarization process,
the binarization process is performed based on a maximum value of the information on the quantized escape value, and
the maximum value is derived based on a quantization parameter derived from the information on the quantization parameter,
wherein the binarization process includes performing a truncated binary (TB) binarization on the information on the quantized escape value based on the maximum value,
wherein the information on the quantized escape value has symbol values from 0 to the maximum value,
in the TB binarization on the information on the quantized escape value, first u symbols are indicated based on k bits and remaining cMax−u+1 symbols are indicated based on (k+1) bits,
the cMax indicates the maximum value,
the k indicates an integer satisfying 2 k ≤cMax<2 k+1 , and
the u is derived based on 2 k+1 −(cMax+1).

2. The image decoding method of claim 1 , wherein,
in the TB binarization, the maximum value is derived based on Clip(0,(1<<bitDepth) 1, (quantScale[qRem]*maxValue+(1<<(qBits−1)))>>qBits, bitDepth is a bit depth, quantScale is an array including k scale values, max Value is derived based on (1<<bitDepth) −1, qRem is derived based on QP % 6, qBits is derived based on 14+(QP/6), and QP is the quantization parameter.

3. The image decoding method of claim 1 , wherein
the binarization process includes performing fixed length binarization on information on the quantized escape value.

4. The image decoding method of claim 3 , wherein,
in the fixed length binarization, the maximum value is derived based on Clip(0, (1<<bitDepth)−1, (1<<binLength)−1),
bitDepth is a bit depth,
binLength is derived based on bitDepth−(Max(0,QP−5)/6), and
QP is the quantization parameter.

5. The image decoding method of claim 1 , wherein
the binarization process includes performing k-th order Exp-Golomb (EG) binarization on the information on the quantized escape information.

6. The image decoding method of claim 5 , wherein,
in the k-th order EG binarization, the k-th order is determined variably based on the maximum value.

7. The image decoding method of claim 1 , wherein
the information on the quantized escape value is a syntax element for a quantized escape value of an escape-coded sample in the current block.

8. The image decoding method of claim 1 , wherein
the information on the quantization parameter is a syntax element for a minimum quantization parameter for a transform skip mode.

9. An image encoding method performed by an encoding apparatus, the image encoding method comprising:
deriving an escape value in a palette mode for a current block;
deriving a quantized escape value based on the escape value and a quantization parameter; and
encoding image information including information on the quantized escape value,
wherein the information on the quantized escape value is encoded based on a binarization process,
the binarization process is performed based on a maximum value of the information on the quantized escape value, and
the maximum value is derived based on the quantization parameter,
wherein the binarization process includes performing a truncated binary (TB) binarization on the information on the quantized escape value based on the maximum value,
wherein the information on the quantized escape value has symbol values from 0 to the maximum value,
in the TB binarization on the information on the quantized escape value, first u symbols are indicated based on k bits and remaining cMax−u+1 symbols are indicated based on (k+1) bits,
the cMax indicates the maximum value,
the k indicates an integer satisfying 2 k ≤cMax <2 k+1 , and
the u is derived based on 2 k+1 −(cMax+1).

10. The image encoding method of claim 9 , wherein
the binarization process includes performing fixed length binarization on information on the quantized escape value.

11. The image encoding method of claim 9 , wherein
the binarization process includes performing k-th order Exp-Golomb (EG) binarization on the information on the quantized escape information.

12. The image encoding method of claim 11 , wherein,
in the k-th order EG binarization, the k-th order is determined variably based on the maximum value.

13. The image encoding method of claim 9 , wherein
the information on the quantized escape value is a syntax element for a quantized escape value of an escape-coded sample in the current block.

14. The image encoding method of claim 9 , wherein
the information on the quantization parameter is a syntax element for a minimum quantization parameter for a transform skip mode.

15. A non-transitory computer-readable storage medium storing an image encoding method generating a bitstream, wherein the image encoding method comprising:
deriving an escape value in a palette mode for a current block;
deriving a quantized escape value based on the escape value and a quantization parameter; and
encoding image information including information on the quantized escape value,
wherein the information on the quantized escape value is encoded based on a binarization process,
the binarization process is performed based on a maximum value of the information on the quantized escape value, and
the maximum value is derived based on the quantization parameter,
wherein the binarization process includes performing a truncated binary (TB) binarization on the information on the quantized escape value based on the maximum value,
wherein the information on the quantized escape value has symbol values from 0 to the maximum value,
in the TB binarization on the information on the quantized escape value, first u symbols are indicated based on k bits and remaining cMax−u+1 symbols are indicated based on (k+1) bits,
the cMax indicates the maximum value,
the k indicates an integer satisfying 2 k ≤cMax<2 k+1 , and
the u is derived based on 2 k+1 −(cMax+1).

Assignees

• Lg Electronics Inc
• Lg Electronics Inc

Inventors

• Yoo Sunmi
• Nam Junghak
• Kim Seunghwan
• Zhao Jie

Classifications

• H04N19/186

  the unit being a colour or a chrominance component · CPC title

• H04N19/176

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

• H04N19/91

  Entropy coding, e.g. variable length coding [VLC] or arithmetic coding · CPC title

• H04N19/13

  Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC] · CPC title