# US 12,223,576 B2 · Apparatus and method for creating avatar

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

- Number: US 12,223,576 B2
- Document ID: US12223576B2
- Status: Granted
- Kind: B2
- Application number: 18/071,640
- Filed: 2022-11-30
- Published: 2024-05-30
- Granted: 2025-02-11
- Earliest claimed: 2022-11-29
- Assignees: Illuni Inc.
- Applicants: illuni Inc.
- Inventors: Byung Hwa Park; Jae Wang Lee; Gabee Jo; Ho Jeong Shin; Kun Sang Jung
- Web page: https://patentgrep.com/patent/US12223576B2

## CPC

- G06T 13/40: of characters, e.g. humans, animals or virtual beings
- G06T 19/00: Manipulating three-dimensional [3D] models or images for computer graphics
- G06T 19/20: Editing of three-dimensional [3D] images, e.g. changing shapes or colours, aligning objects or positioning parts
- G06T 2219/2016: Rotation, translation, scaling
- G06T 2219/2021: Shape modification
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Disclosed is an avatar creation method including creating a base object of an avatar, creating at least one partial object; acquiring customizing information of a user, deforming the base object and the at least one partial object based on the customizing information; and creating a user avatar with a default facial expression by projecting the at least one partial object onto the surface of the base object.

## Claims

### Claim 1

An avatar creation method comprising:
creating a base object of an avatar;
creating at least one partial object;
acquiring customizing information of a user;
deforming the base object and the at least one partial object based on the customizing information;
creating a user avatar with a default facial expression by projecting the at least one partial object onto the surface of the base object; and
changing a facial expression of the user avatar,
wherein the changing the facial expression of the user avatar comprises:
extracting a transformation relationship between the user avatar with the default facial expression and a prepared reference avatar with a default facial expression;
creating the user avatar for each of a plurality of reference facial expressions from the reference avatar prepared for each of the plurality of reference facial expressions based on the transformation relationship; and
changing the facial expression of the user avatar by combining the user avatar with the default facial expression and the user avatar created for each of the plurality of reference facial expressions.

### Claim 2 (depends on 1)

The avatar creation method of claim 1, wherein the deforming the base object and the at least one partial object comprises:
deforming information of the base object included in a grid space through deformation of at least one of a plurality of grid spaces that constitutes the base object based on first customizing information on the base object; and
deforming information of the at least one partial object included in a grid space through deformation of at least one of a plurality of grid spaces that constitutes the at least one partial object based on second customizing information on the at least one partial object.

### Claim 3 (depends on 2)

The avatar creation method of claim 2, wherein the deforming the base object and the at least one partial object further comprises changing at least one of a thickness, a color, a texture, and smoothing of the at least one partial object based on the second customizing information on the at least one partial object.

### Claim 4 (depends on 2)

The avatar creation method of claim 2, wherein a change in a location of a vertex associated with configuration information of the base object accompanies deformation of a partial object associated with the vertex of which the location is changed.

### Claim 5 (depends on 2)

The avatar creation method of claim 2, wherein deformation of at least one of the plurality of grid spaces that constitutes the base object accompanies deformation of a grid space of a partial object associated with the deformation of the grid space of the base object such that the base object and the partial object are deformed together.

### Claim 6 (depends on 1)

The avatar creation method of claim 1, wherein the at least one partial object is projected onto the surface of the base object in a predetermined one direction.

### Claim 7 (depends on 1)

The avatar creation method of claim 1, wherein the at least one partial object is projected in a normal vector direction of the surface of the base object at a location at which the at least one partial object is projected onto the surface of the base object.

### Claim 8 (depends on 1)

The avatar creation method of claim 1, further comprising:
acquiring a captured face image of the user,
wherein the base object and the at least one partial object of the avatar are created based on feature information acquired from the captured face image of the user.

### Claim 9 (depends on 8)

The avatar creation method of claim 8, wherein the feature information acquired from the captured face image of the user includes at least one of a face shape of the user, a skin tone of the user, a hair style of the user, a hair color of the user, an eye size of the user, an eye color of the user, gender of the user, age of the user, a type and a shape of accessory, and a racial characteristic of the user.

### Claim 10 (depends on 1)

The avatar creation method of claim 1, wherein the extracting the transformation relationship comprises:
extracting a first transformation relationship between a base object of the user avatar and a base object of the reference avatar from the default facial expression; and
extracting a second transformation relationship between at least one partial object of the user avatar and a partial object of the reference avatar from the default facial expression, and the creating the user avatar for each of the plurality of reference facial expressions comprises:
creating the base object of the user avatar for each of the plurality of reference facial expressions from the base object of the reference avatar prepared for each of the plurality of reference facial expressions based on the first transformation relationship; and
creating the partial object of the user avatar for each of the plurality of reference facial expressions from the base object of the reference avatar prepared for each of the plurality of reference facial expressions based on the second transformation relationship.

### Claim 11 (depends on 1)

The avatar creation method of claim 1, wherein the changing the facial expression of the user avatar comprises:
acquiring speech information of the user;
determining a combination weight of each of the plurality of reference facial expressions based on the speech information of the user; and
combining the user avatar with the default facial expression and the user avatar created for each of the plurality of reference facial expressions based on the combination weight.

### Claim 12 (depends on 11)

The avatar creation method of claim 11, wherein a combination weight of the user avatar created for each of the plurality of reference facial expressions is expressed as a time-series weight sequence, and
a facial expression of the user avatar is animated based on the time-series weight sequence.

### Claim 13 (depends on 1)

The avatar creation method of claim 1, further comprising:
providing the user avatar as three-dimensional (3D) printing data.

### Claim 14

An avatar creation apparatus comprising:
at least one processor,
wherein the at least one processor is configured to perform a process of creating a base object of an avatar; a process of creating at least one partial object; a process of acquiring customizing information of a user; a process of deforming the base object and the at least one partial object based on the customizing information; a process of creating a user avatar with a default facial expression by projecting the at least one partial object onto the surface of the base object; and a process of changing a facial expression of the user avatar,
wherein the process of changing the facial expression of the user avatar comprises:
a process of extracting a transformation relationship between the user avatar with the default facial expression and a prepared reference avatar with a default facial expression;
a process of creating the user avatar for each of a plurality of reference facial expressions from the reference avatar prepared for each of the plurality of reference facial expressions based on the transformation relationship; and
a process of changing the facial expression of the user avatar by combining the user avatar with the default facial expression and the user avatar created for each of the plurality of reference facial expressions.