# US 2024/0205401 A1 · Low-frequency non-separable transform design method and device

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0205401 A1
- Document ID: US20240205401A1
- Status: Published application
- Kind: A1
- Application number: 18/286,683
- Filed: 2022-04-12
- Published: 2024-06-20
- Earliest claimed: 2021-04-12
- Assignees: LG Electronics Inc.
- Applicants: LG Electronics Inc.
- Inventors: Moonmo KOO; Seunghwan KIM; Jaehyun LIM
- Web page: https://patentgrep.com/patent/US20240205401A1

## CPC

- H04N 19/12: Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
- H04N 19/132: Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking
- H04N 19/124: Quantisation
- H04N 19/126: Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
- H04N 19/167: Position within a video image, e.g. region of interest [ROI]
- H04N 19/176: the region being a block, e.g. a macroblock
- H04N 19/18: the unit being a set of transform coefficients
- H04N 19/60: using transform coding
- H04N 19/61: in combination with predictive coding
- H04N 19/70: characterised by syntax aspects related to video coding, e.g. related to compression standards
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

According to the disclosure of the present document, a quantization parameter for a quantized transform coefficient present at a DC position within a current block can have the smallest value among quantization parameters which are applied to quantized transform coefficients within the current block, on the basis that an LFNST is applied to the current block. Therefore, coding performance capable of being brought by the LFNST can be maximized within the implementation complexity permitted in forthcoming standards.

## Claims

### Claim 1

An image decoding method performed by a decoding apparatus, the method comprising:
obtaining image information through a bitstream, wherein the image information includes residual information and information related to a quantization parameter;
deriving quantized transform coefficients for a current block based on the residual information;
deriving transform coefficients for the current block based on the quantized transform coefficients and the information related to the quantization parameter;
deriving modified transform coefficients by applying a low-frequency non-separable transform (LFNST) based on the transform coefficients;
generating residual samples for the current block based on the modified transform coefficients; and
generating reconstructed samples for the current block based on the residual samples,
wherein based on the LFNST being applied to the current block, a quantization parameter for a quantized transform coefficient present at a DC position in the current block has a smallest value among quantization parameters applied to the quantized transform coefficients in the current block.

### Claim 2 (depends on 1)

The image decoding method of claim 1,
wherein, among the quantized transform coefficients in the current block, the quantization parameter for the quantized transform coefficient present at the DC position has a value of Q−A, and quantization parameters for remaining quantized transform coefficients have a value of Q.

### Claim 3 (depends on 2)

The image decoding method of claim 2,
wherein the information related to the quantization parameter includes information related to a value of A, and
wherein the information related to the value of A is included in at least one of a sequence parameter set (SPS), a picture parameter set (PPS), a picture header (PH), or a slice header (SH).

### Claim 4 (depends on 1)

The image decoding method of claim 1,
wherein coordinates of positions of the quantized transform coefficients in the current block are (x,y) based on that coordinates of the DC position in the current block are (0, 0), an x-coordinate increases from left to right, and an y-coordinate increases from top to bottom,
wherein the quantization parameters for the quantized transform coefficients in the current block have a value of Q+As, and
wherein a value of As is determined based on a value of x+y.

### Claim 5 (depends on 4)

The image decoding method of claim 4,
wherein the value of As is −1 I based on a case where the value of x+y is 0, and
wherein the value of As is 0 based on a case where the value of x+y is not 0.

### Claim 6 (depends on 4)

The image decoding method of claim 4,
wherein the value of As is −2 based on a case where the value of x+y is 0.
wherein the value of As is −1 based on a case where the value of x+y is 1 or 2,
wherein the value of As is 0 based on a case where the value of x+y is not one of 0, 1 or 2.

### Claim 7 (depends on 4)

The image decoding method of claim 4,
wherein the information related to the quantization parameter includes information related to a value of As, and
wherein the information related to the value of As is included in at least one of a sequence parameter set (SPS), a picture parameter set (PPS), a picture header (PH), or a slice header (SH).

### Claim 8 (depends on 1)

The image decoding method of claim 1,
wherein the information related to the quantization parameter includes information related to a scaling list, and
wherein based on the LFNST being applied to the current block, the scaling list is not applied to the current block.

### Claim 9 (depends on 1)

The image decoding method of claim 1,
wherein based on the LFNST being not applied to the current block, a first scaling list is applied to the quantized transform coefficients to derive the transform coefficients,
wherein based on the LFNST being applied to the current block, a second scaling list is applied to the quantized transform coefficients to derive the transform coefficients, and
wherein the first scaling list and the second scaling list are different.

### Claim 10 (depends on 9)

The image decoding method of claim 9,
wherein the second scaling list is configured based on at least one of positions where LFNST transform coefficients exist in the current block, a size of the current block, or an LFNST kernel.

### Claim 11

An image encoding method performed by an encoding apparatus, the method comprising:
generating residual samples for a current block;
deriving transform coefficients based on the residual samples;
deriving modified transform coefficients by applying an low-frequency non-separable transform (LFNST) to the transform coefficients;
deriving quantized transform coefficients for the modified transform coefficients based on the modified transform coefficients and a quantization parameter;
generating information related to the quantization parameter:
generating residual information based on the quantized transform coefficients; and
encoding image information including the residual information and the information related to the quantization parameter.
wherein based on the LFNST being applied to the current block, a quantization parameter for a modified transform coefficient present at a DC position in the current block has a smallest value among quantization parameters applied to the modified transform coefficients in the current block.

### Claim 12 (depends on 11)

The image encoding method of claim 11,
wherein, among the quantized transform coefficients in the current block, the quantization parameter for the modified transform coefficient present at the DC position has a value of Q−A, and quantization parameters for remaining modified transform coefficients have a value of Q.

### Claim 13 (depends on 12)

The image encoding method of claim 12,
wherein the information related to the quantization parameter includes information related to a value of A, and
wherein the information related to the value of A is included in at least one of a sequence parameter set (SPS), a picture parameter set (PPS), a picture header (PH), or a slice header (SH).

### Claim 14 (depends on 11)

The image encoding method of claim 11,
wherein coordinates of positions of the quantized transform coefficients in the current block are (x,y) based on that coordinates of the DC position in the current block are (0, 0), an x-coordinate increases from left to right, and an y-coordinate increases from top to bottom.
wherein the quantization parameters for the modified transform coefficients in the current block have a value of Q+As, and
wherein a value of As is determined based on a value of x+y.

### Claim 15

A transmission method of data for an image, the method comprising:
obtaining a bitstream for the image, wherein the bitstream is generated based on generating residual samples for a current block, deriving transform coefficients based on the residual samples, deriving modified transform coefficients by applying an low-frequency non-separable transform (LFNST) to the transform coefficients, deriving quantized transform coefficients for the modified transform coefficients based on the modified transform coefficients and a quantization parameter, generating information related to the quantization parameter, generating residual information based on the quantized transform coefficients, and encoding image information including the residual information and the information related to the quantization parameter; and
transmitting the data comprising the bitstream,
wherein based on the LFNST being applied to the current block, a quantization parameter for a modified transform coefficient present at a DC position in the current block has a smallest value among quantization parameters applied to the modified transform coefficients in the current block.
