# US 2023/0069310 A1 · Object classification using one or more neural networks

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

- Number: US 2023/0069310 A1
- Document ID: US20230069310A1
- Status: Published application
- Kind: A1
- Application number: 17/398,655
- Filed: 2021-08-10
- Published: 2023-03-02
- Earliest claimed: 2021-08-10
- Assignees: NVIDIA Corporation
- Applicants: Nvidia Corporation
- Inventors: Andriy Myronenko; Ziyue Xu; Dong Yang; Holger Roth; Daguang Xu
- Web page: https://patentgrep.com/patent/US20230069310A1

## CPC

- G06T 7/11: Region-based segmentation
- G06V 10/82: using neural networks
- G06F 18/2451: linear, e.g. hyperplane
- G06N 3/08: Learning methods
- G06T 2207/20021: Dividing image into blocks, subimages or windows
- G06T 2207/20084: Artificial neural networks [ANN]
- G06V 2201/03: Recognition of patterns in medical or anatomical images
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Apparatuses, systems, and techniques are presented to classify objects in images. In at least one embodiment, one or more neural networks are used to identify one or more objects in one or more full images based, at least in part, on the one or more neural networks having been trained using the one or more full images and one or more portions of the one or more full images.

## Claims

### Claim 1

A processor, comprising:
one or more circuits to train one or more neural networks based, at least in part, on one or more full digital images and one or more portions of the one or more full digital images.

### Claim 2 (depends on 1)

The processor of claim 1, wherein the one or more neural networks are trained to account for dependencies between the one or more portions of the one or more full digital images.

### Claim 3 (depends on 2)

The processor of claim 2, wherein the one or more neural networks include a neural network with one or more embedded transformer encoder blocks to capture the dependencies between the one or more portions.

### Claim 4 (depends on 1)

The processor of claim 1, wherein the one or more neural networks are to determine pseudo-labels for the one or more portions.

### Claim 5 (depends on 4)

The processor of claim 4, wherein the one or more circuits are to further train the one or more neural networks using image-level labels for the one or more full digital images and the pseudo-labels for the one or more portions.

### Claim 6 (depends on 4)

The processor of claim 4, wherein the one or more networks are trained using instance-wise loss supervision based, at least in part, upon the pseudo-labels.

### Claim 7

A system comprising:
one or more processors to use one or more neural networks to identify one or more objects in one or more full images based, at least in part, on the one or more neural networks having been trained using the one or more full images and one or more portions of the one or more full images.

### Claim 8 (depends on 7)

The system of claim 7, wherein the one or more neural networks are trained to account for dependencies between the one or more portions of the one or more full digital images.

### Claim 9 (depends on 8)

The system of claim 8, wherein the one or more neural networks include a neural network with one or more embedded transformer encoder blocks to capture the dependencies between the one or more portions.

### Claim 10 (depends on 7)

The system of claim 7, wherein the one or more neural networks are to determine pseudo-labels for the one or more portions.

### Claim 11 (depends on 10)

The system of claim 10, wherein the one or more processors are to further train the one or more neural networks using image-level labels for the one or more full digital images and the pseudo-labels for the one or more portions.

### Claim 12 (depends on 10)

The system of claim 10, wherein the one or more networks are trained using instance-wise loss supervision based, at least in part, upon the pseudo-labels.

### Claim 13

A method comprising:
training one or more neural networks based, at least in part, on one or more full digital images and one or more portions of the one or more full digital images.

### Claim 14 (depends on 13)

The method of claim 13, wherein the one or more neural networks are trained to account for dependencies between the one or more portions of the one or more full digital images.

### Claim 15 (depends on 14)

The method of claim 14, wherein the one or more neural networks include a neural network with one or more embedded transformer encoder blocks to capture the dependencies between the one or more portions.

### Claim 16 (depends on 13)

The method of claim 13, wherein the one or more neural networks are to determine pseudo-labels for the one or more portions.

### Claim 17 (depends on 16)

The method of claim 16, further comprising:
further training the one or more neural networks using image-level labels for the one or more full digital images and the pseudo-labels for the one or more portions.

### Claim 18 (depends on 16)

The method of claim 16, wherein the one or more networks are trained using instance-wise loss supervision based, at least in part, upon the pseudo-labels.

### Claim 19

A machine-readable medium having stored thereon a set of instructions, which if performed by one or more processors, cause the one or more processors to at least:
use one or more neural networks to identify one or more objects in one or more full images based, at least in part, on the one or more neural networks having been trained using the one or more full images and one or more portions of the one or more full images.

### Claim 20 (depends on 19)

The machine-readable medium of claim 19, wherein the one or more neural networks are trained to account for dependencies between the one or more portions of the one or more full digital images.

### Claim 21 (depends on 20)

The machine-readable medium of claim 20, wherein the one or more neural networks include a neural network with one or more embedded transformer encoder blocks to capture the dependencies between the one or more portions.

### Claim 22 (depends on 19)

The machine-readable medium of claim 19, wherein the one or more neural networks are to determine pseudo-labels for the one or more portions.

### Claim 23 (depends on 22)

The machine-readable medium of claim 22, wherein the instructions if performed further cause the one or more processors to:
further train the one or more neural networks using image-level labels for the one or more full digital images and the pseudo-labels for the one or more portions.

### Claim 24 (depends on 22)

The machine-readable medium of claim 22, wherein the one or more networks are trained using instance-wise loss supervision based, at least in part, upon the pseudo-labels.

### Claim 25

An image classification system, comprising:
one or more processors to use one or more neural networks to identify one or more images in one or more full images based, at least in part, on the one or more neural networks having been trained using the one or more full images and one or more portions of the one or more full images; and
memory for storing network parameters for the one or more neural networks.

### Claim 26 (depends on 25)

The image classification system of claim 25, wherein the one or more neural networks are trained to account for dependencies between the one or more portions of the one or more full digital images.

### Claim 27 (depends on 26)

The image classification system of claim 26, wherein the one or more neural networks include a neural network with one or more embedded transformer encoder blocks to capture the dependencies between the one or more portions.

### Claim 28 (depends on 25)

The image classification system of claim 25, wherein the one or more neural networks are to determine pseudo-labels for the one or more portions.

### Claim 29 (depends on 28)

The image classification system of claim 28, wherein the one or more processors are to further train the one or more neural networks using image-level labels for the one or more full digital images and the pseudo-labels for the one or more portions.

### Claim 30 (depends on 28)

The image classification system of claim 28, wherein the one or more networks are trained using instance-wise loss supervision based, at least in part, upon the pseudo-labels.
