# US 2025/0168331 A1 · Intra prediction 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 2025/0168331 A1
- Document ID: US20250168331A1
- Status: Published application
- Kind: A1
- Application number: 19/030,269
- Filed: 2025-01-17
- Published: 2025-05-22
- Earliest claimed: 2016-06-22
- Assignees: Lx Semicon Co., Ltd.
- Applicants: LX Semicon Co., Ltd.
- Inventors: Hyun Suk KO; Jin Ho LEE; Sung Chang LIM; Jung Won KANG; Ha Hyun LEE; Dong San JUN; Seung Hyun CHO; Hui Yong KIM; Jin Soo CHOI
- Web page: https://patentgrep.com/patent/US20250168331A1

## CPC

- H04N 19/11: among a plurality of spatial predictive coding modes
- H04N 19/105: Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
- H04N 19/159: Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
- H04N 19/176: the region being a block, e.g. a macroblock
- H04N 19/184: the unit being bits, e.g. of the compressed video stream
- H04N 19/593: involving spatial prediction techniques
- H04N 19/96: Tree coding, e.g. quad-tree coding
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

An image decoding/encoding method and apparatus are discussed. The image decoding method can include obtaining, through a bitstream, image information comprising information on quantized transform coefficient levels, deriving transform coefficients of a current block based on the information on quantized transform coefficient levels, and determining a transform set for a secondary inverse-transform based on an intra prediction mode of the current block.

## Claims

### Claim 1

An image decoding method performed by an image decoding apparatus, the method comprising:
obtaining, through a bitstream, image information comprising information on quantized transform coefficient levels;
deriving transform coefficients of a current block based on the information on quantized transform coefficient levels;
determining a transform set for a secondary inverse-transform based on an intra prediction mode of the current block;
determining a transform kernel for the secondary inverse-transform based on the transform set and a size of the current block;
deriving modified transform coefficients by performing the secondary inverse-transform based on the transform coefficients and the transform kernel for the secondary inverse-transform;
deriving a reconstructed residual signal by performing a primary inverse-transform based on the modified transform coefficients; and
generating a reconstructed current block based on the reconstructed residual signal,
wherein the secondary inverse-transform is a non-separable inverse-transform,
wherein the transform set including plural transform kernel candidates is determined based on the intra prediction mode of the current block,
wherein transform kernel candidates in a first transform set is different from transform kernel candidates in a second transform set,
wherein one of the plural transform kernel candidates in the transform set is determined as the transform kernel for the secondary inverse-transform based on the size of the current block,
wherein a first transform kernel determined from the transform set based on a first block size of the current block is different from a second transform kernel determined from the transform set based on a second block size of the current block.

### Claim 2 (depends on 1)

The image decoding method of claim 1, wherein a number of the plural transform kernel candidates in the transform set is determined differently based on the size of the current block.

### Claim 3

An image encoding method performed by an image encoding apparatus, the method comprising:
deriving a residual signal for a current block;
deriving transform coefficients by performing a primary transform based on the residual signal;
determining a transform set for a secondary transform based on an intra prediction mode of the current block;
determining a transform kernel for the secondary transform based on the transform set and a size of the current block;
deriving modified transform coefficients by performing the secondary transform on the transform coefficients based on the transform kernel for the secondary transform;
deriving quantized transform coefficient levels based on the modified transform coefficients; and
encoding image information including information on the quantized transform coefficient levels,
wherein the secondary transform is a non-separable transform,
wherein the transform set including plural transform kernel candidates is determined based on the intra prediction mode of the current block,
wherein transform kernel candidates in a first transform set is different from transform kernel candidates in a second transform set,
wherein one of the plural transform kernel candidates in the transform set is determined as the transform kernel for the secondary inverse-transform based on the size of the current block,
wherein a first transform kernel determined from the transform set based on a first block size of the current block is different from a second transform kernel determined from the transform set based on a second block size of the current block.

### Claim 4 (depends on 3)

The image encoding method of claim 3, wherein a number of the plural transform kernel candidates in the transform set is determined differently based on the size of the current block.

### Claim 5

A transmission method for image data, the method comprising:
obtaining a bitstream of encoded image information, wherein the encoded image information is generated based on deriving a residual signal for a current block, deriving transform coefficients by performing a primary transform based on the residual signal, determining a transform set for a secondary transform based on an intra prediction mode of the current block, determining a transform kernel for the secondary transform based on the transform set and a size of the current block, deriving modified transform coefficients by performing the secondary transform based on the transform coefficients and the transform kernel for the secondary transform, deriving quantized transform coefficient levels based on the modified transform coefficients, and encoding image information including information on the quantized transform coefficient levels; and
transmitting the image data comprising the bitstream,
wherein the secondary transform is a non-separable transform,
wherein the transform set including plural transform kernel candidates is determined based on the intra prediction mode of the current block,
wherein transform kernel candidates in a first transform set is different from transform kernel candidates in a second transform set,
wherein one of the plural transform kernel candidates in the transform set is determined as the transform kernel for the secondary inverse-transform based on the size of the current block,
wherein a first transform kernel determined from the transform set based on a first block size of the current block is different from a second transform kernel determined from the transform set based on a second block size of the current block.

### Claim 6 (depends on 5)

The method of claim 5, wherein a number of the plural transform kernel candidates in the transform set is determined differently based on the size of the current block.