# US 2023/0145028 A1 · Method and apparatus for feature transforming and processing based on artificial neural network

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

- Number: US 2023/0145028 A1
- Document ID: US20230145028A1
- Status: Published application
- Kind: A1
- Application number: 17/939,615
- Filed: 2022-09-07
- Published: 2023-05-11
- Earliest claimed: 2021-11-10
- Assignees: Electronics and Telecommunications Research Institute
- Applicants: Electronics and Telecommunications Research Institue
- Inventors: Sung Chan OH; Yong Jin KWON; Hyung Il KIM; Jin Young MOON; Yu Seok BAE; Ki Min YUN; Jeun Woo LEE; Joong Won HWANG
- Web page: https://patentgrep.com/patent/US20230145028A1

## CPC

- G06V 10/82: using neural networks
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/045: Combinations of networks
- G06N 3/048: Activation functions
- G06N 3/08: Learning methods
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06V 10/454: {Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]}
- G06V 10/7715: {Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods}
- G06V 10/84: using probabilistic graphical models from image or video features, e.g. Markov models or Bayesian networks
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Disclosed herein are a method and apparatus for processing feature information based on an artificial neural network. According to an embodiment of the present disclosure, the apparatus for processing feature information based on an artificial neural network may include a memory for storing data and a processor for controlling the memory, and the processor may further be configured to extract a graph, which includes vertices, based on a feature map of an image, to extract a feature vector corresponding to the vertices and to process the graph and the feature vector based on an artificial neural network, and the graph may include positions of the vertices and information on a connection relationship between the vertices.

## Claims

### Claim 1

An apparatus for processing feature information based on an artificial neural network, the apparatus comprising:
a memory configured to store data; and
a processor configured to control the memory,
wherein the processor is further configured to:
extract a graph, which includes vertices, based on a feature map of an image,
extract a feature vector corresponding to the vertices, and
process the graph and the feature vector based on an artificial neural network, and
wherein the graph includes positions of the vertices and information on a connection relationship between the vertices.

### Claim 2 (depends on 1)

The apparatus of claim 1, wherein the artificial neural network is a graph convolution network (GCN).

### Claim 3 (depends on 1)

The apparatus of claim 1, wherein the information on the connection relationship between the vertices includes an edge set and an adjacency matrix for the vertices.

### Claim 4 (depends on 3)

The apparatus of claim 3, wherein the edge set and the adjacency matrix are be obtained using a k-nearest neighbors algorithm, a random sampling algorithm, or a Delaunay triangulation algorithm.

### Claim 5 (depends on 1)

The apparatus of claim 1, wherein the feature vector is obtained through a graph convolution layer.

### Claim 6 (depends on 5)

The apparatus of claim 5, wherein the feature vector is obtained using a k-nearest neighbors algorithm or a random sampling algorithm.

### Claim 7 (depends on 1)

The apparatus of claim 1, wherein the vertices are obtained using the feature map and an importance map for the feature map, which is obtained through a convolution layer.

### Claim 8

A method for processing feature information based on an artificial neural network, the method comprising:
extracting a graph consisting of vertices based on a feature map of an image;
extracting a feature vector corresponding to the vertices; and
processing the graph and the feature vector based on an artificial neural network,
wherein the graph includes positions of the vertices and information on a connection relationship between the vertices.

### Claim 9 (depends on 8)

The method of claim 8, wherein the artificial neural network is a graph convolution network (GCN).

### Claim 10 (depends on 8)

The method of claim 8, wherein the information on the connection relationship between the vertices includes an edge set and an adjacency matrix for the vertices.

### Claim 11 (depends on 10)

The method of claim 10, wherein the edge set and the adjacency matrix are be obtained using a k-nearest neighbors algorithm, a random sampling algorithm, or a Delaunay triangulation algorithm.

### Claim 12 (depends on 8)

The method of claim 8, wherein the feature vector is obtained through a graph convolution layer.

### Claim 13 (depends on 12)

The method of claim 12, wherein the feature vector is obtained using a k-nearest neighbors algorithm or a random sampling algorithm.

### Claim 14 (depends on 8)

The method of claim 8, wherein the vertices are obtained using the feature map and an importance map for the feature map, which is obtained through a convolution layer.

### Claim 15

An apparatus for processing feature information based on an artificial neural network, the apparatus comprising:
a memory configured to store data; and
a processor configured to control the memory,
wherein the processor is further configured to:
obtain a graph including vertices, and a feature vector, which are processed based on an artificial neural network,
output a feature map of an image based on the graph and the feature vector, and
obtain a per-pixel analysis result of the image based on the feature map, and
wherein the graph includes positions of the vertices and information on a connection relationship between the vertices.

### Claim 16 (depends on 15)

The apparatus of claim 15, wherein the artificial neural network is a graph convolution network (GCN).

### Claim 17 (depends on 15)

The apparatus of claim 15, wherein the output feature map of the image is output through a convolution layer.

### Claim 18 (depends on 15)

The apparatus of claim 15, wherein the per-pixel analysis result is obtained through a fully-connected layer.
