# US 12,223,603 B2 · Method of learning a target object using a virtual viewpoint camera and a method of augmenting a virtual model on a real object implementing the target object using the same

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

- Number: US 12,223,603 B2
- Document ID: US12223603B2
- Status: Granted
- Kind: B2
- Application number: 17/991,791
- Filed: 2022-11-21
- Published: 2023-06-29
- Granted: 2025-02-11
- Earliest claimed: 2021-12-29
- Assignees: Virnect Inc.
- Applicants: VIRNECT inc.
- Inventors: Ki Young Kim; Thorsten Korpitsch
- Web page: https://patentgrep.com/patent/US12223603B2

## CPC

- G06T 19/006: {Mixed reality (object pose determination, tracking or camera calibration for mixed reality G06T7/00)}
- G06V 10/00: Arrangements for image or video recognition or understanding (character recognition in images or video G06V30/10)
- G06T 15/20: Perspective computation
- G06T 19/20: Editing of three-dimensional [3D] images, e.g. changing shapes or colours, aligning objects or positioning parts
- G06V 10/74: Image or video pattern matching; Proximity measures in feature spaces
- G06V 20/20: in augmented reality scenes
- G06T 2215/16: Using real world measurements to influence rendering
- G06T 2219/2016: Rotation, translation, scaling
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Provided is a method of learning a target object implemented on a computer-aided design program of an authoring computing device using a virtual viewpoint camera, including displaying a digital model of a target object that is a target for image recognition, setting at least one observation area surrounding the digital model of the target object and having a plurality of viewpoints on the digital model, generating a plurality of pieces of image data obtained by viewing the digital model of the target object at the plurality of viewpoints of the at least one observation area, and generating object recognition library data for recognizing a real object implementing the digital model of the target object based on the generated plurality of pieces of image data.

## Claims

### Claim 1

A method of learning a target object implemented on a computer-aided design program of an authoring computing device using a virtual viewpoint camera, comprising:
displaying a digital model of a target object that is a target for image recognition;
setting at least one observation area surrounding the digital model of the target object and having a plurality of viewpoints on the digital model, wherein the viewpoints include a plurality of active viewpoints and a plurality of inactive viewpoints;
generating a plurality of pieces of image data obtained by viewing the digital model of the target object at the plurality of viewpoints of the at least one observation area; and
generating object recognition library data for recognizing a real object implementing the digital model of the target object based on the generated plurality of pieces of image data,
wherein the observation area has a hemispherical shape, and the viewpoints closest to each other among the plurality of viewpoints are spaced apart at regular intervals,
wherein an observation viewpoint of the digital model is set by converting at least some of the inactive viewpoints into the active viewpoints or converting all inactive viewpoints into the active viewpoints, wherein the all inactive viewpoints are equally adjusted the distances between the plurality of inactive points.

### Claim 2 (depends on 1)

The method of claim 1, wherein the generating the plurality of pieces of image data comprises generating a plurality of pieces of image data obtained by viewing the digital model of the target object at the plurality of viewpoints in a direction of an origin of the hemispherical shape.

### Claim 3 (depends on 1)

The method of claim 1, wherein the setting the observation area comprises setting a plurality of hemispherical observation areas each having the same origin and different radii in order to observe the digital model of the target object by distance.

### Claim 4 (depends on 3)

The method of claim 3, wherein the setting the observation area comprises rotating any one of the plurality of hemispherical observation areas around the origin in order to observe the digital model of the target object at different viewpoints by distance.

### Claim 5 (depends on 1)

The method of claim 1, further comprising changing a position of the digital model of the target object with respect to the observation area.

### Claim 6 (depends on 5)

The method of claim 5, further comprising allowing the digital model to be observed only at the active viewpoints among the plurality of viewpoints of the observation area.

### Claim 7 (depends on 2)

The method of claim 2, further comprising adjusting a magnification of the target object based on distances from the viewpoints to the origin such that the digital model of the target object has the same size as a size of the real object when the distance from a shooting point to the digital model at a viewpoint is the same as the distance from a camera of a user computing device to the real object.

### Claim 8 (depends on 1)

The method of claim 1, further comprising generating augmented content, matching the augmented content to the digital model of the target object, and storing the augmented content.

### Claim 9 (depends on 1)

A method of augmenting a virtual model on a real object implementing a trained target object using a virtual viewpoint camera, performed by an augmented reality program of a terminal including a camera, the method comprising:
receiving and storing the generated object recognition library data of claim 1;
acquiring a captured image by photographing a surrounding environment;
detecting a real object matching the stored object recognition library data in the acquired captured image; and
displaying augmented content on the detected real object such that the augmented content matches the real object.