# US 12,114,015 B2 · Methods and apparatus for restricted secondary transform and signaling thereof in image coding

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

- Number: US 12,114,015 B2
- Document ID: US12114015B2
- Status: Granted
- Kind: B2
- Application number: 17/783,954
- Filed: 2020-12-10
- Published: 2022-12-01
- Granted: 2024-10-08
- Earliest claimed: 2019-12-12
- Assignees: Hfi Innovation Inc.
- Applicants: HFI Innovation Inc.
- Inventors: Man-Shu Chiang; Tzu-Der Chuang; Chih-Wei Hsu; Ching-Yeh Chen; Zhi-Yi Lin
- Web page: https://patentgrep.com/patent/US12114015B2

## CPC

- H04N 19/157: Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
- H04N 19/70: characterised by syntax aspects related to video coding, e.g. related to compression standards
- 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 19/176: the region being a block, e.g. a macroblock
- H04N 19/186: the unit being a colour or a chrominance component
- H04N 19/96: Tree coding, e.g. quad-tree coding
- 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: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

In a method and apparatus for video coding using Low-Frequency Non-Separable Transform (LFNST) mode, a coding unit (CU) is partitioned into one or more transform blocks (TBs). A syntax is determined at an encoder side or at a decoder side, where the determining step is performed by signaling the syntax at the encoder side or by parsing the syntax at the decoder side if one or more conditions are satisfied. The syntax indicates whether the LFNST mode is applied to the current CU and/or which LFNST kernel is applied when the LFNSF is applied, and the conditions comprise a target condition corresponding to that all target TBs in a target TB set have a TS mode indication as false, and the target TB set is selected from the TBs in the current CU. The current CU is encoded or decoded according to the LFNST mode.

## Claims

### Claim 1

A method for encoding or decoding a video sequence, wherein a Low-Frequency Non-Separable Transform (LFNST) mode and Transform Skip (TS) mode are supported, the method comprising:
receiving input data related to a current coding unit (CU) in a current picture, wherein the CU is partitioned into one or more transform blocks (TBs) and the input data correspond to residual data at a video encoder side or a coded data of the current CU at a video decoder side;
determining a syntax at an encoder side or at a decoder side, wherein said determining the syntax is performed by signaling the syntax at the encoder side or by parsing the syntax at the decoder side if one or more conditions are satisfied, wherein the syntax indicates whether the LFNST mode is applied to the current CU and/or which LFNST kernel is applied if the LFNST mode is applied, and said one or more conditions comprise a target condition corresponding to that all target TBs in a target TB set have a TS mode indication among more than one color components as false, and wherein the target TB set is selected from said one or more TBs; and
encoding the current CU at the encoder side or decoding the current CU at the decoder side according to the LFNST mode as indicated by the syntax.

### Claim 2 (depends on 1)

The method of claim 1, wherein in a luma splitting tree, the current CU corresponds to a luma coding block, and the target TB set corresponds to one or more luma TBs.

### Claim 3 (depends on 1)

The method of claim 1, wherein in a chroma splitting tree, the current CU corresponds to one or more chroma coding blocks, and the target TB set corresponds to one or more chroma TBs.

### Claim 4 (depends on 1)

The method of claim 1, wherein in a single splitting tree, the current CU corresponds to one luma coding block and one or more chroma coding blocks, and the target TB set corresponds to one or more luma TBs and one or more chroma TBs.

### Claim 5 (depends on 1)

The method of claim 1, wherein the current CU corresponds to either one luma coding block or one or more chroma coding blocks, and the target TB set corresponds to one or more TBs for all components in the current CU.

### Claim 6 (depends on 1)

The method of claim 1, wherein the target TB set corresponds to one or more first TBs for each coding block in the current CU.

### Claim 7 (depends on 1)

The method of claim 1, wherein a variable is used to record whether to signal or parse the syntax for the LFNST mode.

### Claim 8 (depends on 1)

The method of claim 1, wherein the syntax is inferred as the LFNST mode not applied to the current CU when the TS mode indication is true.

### Claim 9 (depends on 1)

The method of claim 1, wherein the LFNST mode is not used for a TB in the current CU if the syntax designated as LFNST index for the current CU is equal to zero, and wherein the LFNST index with a value greater than zero indicates one of low-frequency transform matrices selected.

### Claim 10

An apparatus for encoding or decoding a video sequence, wherein a Low-Frequency Non-Separable Transform (LFNST) mode and Transform Skip (TS) mode are supported, the apparatus comprising one or more electronic circuits or processors arranged to:
receive input data related to a current coding unit (CU) in a current picture, wherein the CU is partitioned into one or more transform blocks (TBs) and the input data correspond to residual data at a video encoder side or a coded data of the current CU at a video decoder side;
determining a syntax at an encoder side or at a decoder side, wherein said determining the syntax is performed by signaling the syntax at the encoder side or by parsing the syntax at the decoder side if one or more conditions are satisfied, wherein the syntax indicates whether the LFNST mode is applied to the current CU and/or which LFNST kernel is applied if the LFNST mode is applied, and said one or more conditions comprise a target condition corresponding to that all target TBs in a target TB set have a TS mode indication among more than one color components as false, and wherein the target TB set is selected from said one or more TBs; and
encode the current CU at the encoder side or decode the current CU at the decoder side according to the LFNST mode as indicated by the syntax.

### Claim 11 (depends on 10)

The apparatus of claim 10, wherein in a luma splitting tree, the current CU corresponds to a luma coding block, and the target TB set corresponds to one or more luma TBs.

### Claim 12 (depends on 10)

The apparatus of claim 10, wherein in a chroma splitting tree, the current CU corresponds to one or more chroma coding blocks, and the target TB set corresponds to one or more chroma TBs.

### Claim 13 (depends on 10)

The apparatus of claim 10, wherein in a single splitting tree, the current CU corresponds to one luma coding block and one or more chroma coding blocks, and the target TB set corresponds to one or more luma TBs and one or more chroma TBs.

### Claim 14 (depends on 10)

The apparatus of claim 10, wherein the current CU corresponds to either one luma coding block or one or more chroma coding blocks, and the target TB set corresponds to one or more TBs for all components in the current CU.

### Claim 15 (depends on 10)

The apparatus of claim 10, wherein the target TB set corresponds to one or more first TBs for each coding block in the current CU.

### Claim 16 (depends on 10)

The apparatus of claim 10, wherein a variable is used to record whether to signal or parse the syntax for the LFNST mode.

### Claim 17 (depends on 10)

The apparatus of claim 10, wherein the syntax is inferred as the LFNST mode not applied to the current CU when the TS mode indication is true.

### Claim 18 (depends on 10)

The apparatus of claim 10, wherein the LFNST mode is not used for a TB in the current CU if the syntax designated as LFNST index for the current CU is equal to zero, and wherein the LFNST index with a value greater than zero indicates one of low-frequency transform matrices selected.

### Claim 19 (depends on 1)

The method of claim 1, wherein the target condition corresponding to that all target TBs in a target TB set have the TS mode indication as false for each of said more than one color components.