# US 2026/0065514 A1 · Method of encoding/decoding feature

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

- Number: US 2026/0065514 A1
- Document ID: US20260065514A1
- Status: Published application
- Kind: A1
- Application number: 19/311,153
- Filed: 2025-08-27
- Published: 2026-03-05
- Earliest claimed: 2024-08-28
- Assignees: Electronics and Telecommunications Research Institute
- Applicants: Electronics and Telecommunications Research Institute
- Inventors: Youn Hee KIM; JooYoung LEE; Se Yoon JEONG; Jung Won KANG
- Web page: https://patentgrep.com/patent/US20260065514A1

## CPC

- G06T 9/002: {using neural networks}
- G06T 3/4046: using neural networks
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method of encoding a feature according to the present disclosure may include converting a multi-layer feature to a single-layer feature; generating encoding data by transforming the single-layer feature; generating a 2-dimensional (2D) image based on the encoding data; and encoding the 2D image. In this instance, the encoding data may be generated by reducing channels or reducing a resolution of the single-layer feature.

## Claims

### Claim 1

A method of encoding a feature, comprising:
converting a multi-layer feature to a single-layer feature;
generating encoding data by transforming the single-layer feature;
generating a 2-dimensional (2D) image based on the encoding data; and
encoding the 2D image,
wherein the encoding data is generated by reducing channels or reducing a resolution of the single-layer feature.

### Claim 2 (depends on 1)

The method of claim 1, wherein reducing the channels of the single-layer feature represents removing a non-major channel from the single-layer feature.

### Claim 3 (depends on 2)

The method of claim 2, wherein whether a current channel in the single-layer feature is a major channel or not is determined by comparing a component value, corresponding to the current channel, in an importance vector and a threshold value.

### Claim 4 (depends on 3)

The method of claim 3, wherein the importance vector represents a gain unit vector.

### Claim 5 (depends on 4)

The method of claim 4, wherein the gain unit vector is one of a plurality of gain unit vectors, and
wherein each of the plurality of gain unit vectors is learned for a corresponding bitrate.

### Claim 6 (depends on 5)

The method of claim 5, wherein index information indicating one of the plurality of gain unit vectors is encoded.

### Claim 7 (depends on 1)

The method of claim 1, wherein channels in the single-layer feature are rearranged with reference to a reference system, and
wherein information indicating a location of a last major channel among rearranged channels is encoded.

### Claim 8 (depends on 1)

The method of claim 1, wherein reducing the resolution of the single-layer feature represents removing a non-major component in a current channel in the single-layer feature.

### Claim 9 (depends on 8)

The method of claim 8, wherein a predetermined number of components with highest intensity value in the current channel are selected as a major component.

### Claim 10 (depends on 9)

The method of claim 9, wherein mask information to distinguish a major component and a non-major component in the current channel is encoded.

### Claim 11 (depends on 1)

The method of claim 1, wherein the mask information is encoded for each channel in the single-layer feature.

### Claim 12

A method of decoding a feature, comprising:
decoding a 2-dimensional (2D) image;
reconstructing encoding data based on the 2D image;
restoring a single-layer feature based on the encoding data; and
restoring a multi-layer feature based on the single-layer feature,
wherein the single-layer feature is obtained by increasing channels or increasing a resolution of the encoding data.

### Claim 13 (depends on 12)

The method of claim 12, wherein the single-layer feature is obtained by concatenating a non-major channel to the encoding data.

### Claim 14 (depends on 13)

The method of claim 13, wherein the non-major channel is generated by padding the non-major channel with a pre-defined value.

### Claim 15 (depends on 12)

The method of claim 12, wherein increasing the resolution of the encoding data represents reconstructing a non-major component in a current channel in the encoding data, and
wherein a position of the non-major component in the current channel is determined based on mask information.

### Claim 16

A non-transitory computer readable medium storing instructions when executed cause the computer to carry out a method of encoding a feature which comprising:
converting a multi-layer feature to a single-layer feature;
generating encoding data by transforming the single-layer feature;
generating a 2-dimensional (2D) image based on the encoding data; and
encoding the 2D image,
wherein the encoding data is generated by reducing channels or reducing a resolution of the single-layer feature.
