IP Library Granted Patent US 12675934
Granted Patent B2
US 12675934 · App. 18/400,037 · Granted Jul 7, 2026

Avatar generation system, avatar generation method, and recording medium

Inventors: Kiri Inayoshi (Tokyo, JP); Kentaro Nishida (Tokyo, JP)
Assignee: NEC CORPORATION
G06T13/80G06T7/251G06T11/00G06T13/40G06T2207/30196G06T2210/16G06T2210/21
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12675934
App. No.
18/400,037
Granted
Jul 7, 2026
Kind
B2
Abstract

An avatar generation system includes an acquisition unit, an execution unit, a detection unit, and an output unit. The acquisition unit acquires setting data for bones and meshes of an avatar. The execution unit executes processing of changing a posture of the avatar based on the setting data and a defined movement. The detection unit detects a state of penetration between the meshes constituting the avatar based on data in which the posture of the avatar has been changed. The output unit outputs a result of the detection.

Claims (58)

1 . An avatar generation system comprising:

at least one memory storing instructions;

at least one processor configured to access the at least one memory and execute the instructions to:

acquire setting data for bones and meshes of an avatar;

execute processing of changing a posture of the avatar based on the setting data and a defined movement;

detect a state of penetration between the meshes constituting the avatar based on data in which the posture of the avatar has been changed;

estimate correction amounts of setting values of the meshes based on the result of the detection; and

output a plurality of images of the avatar different from each other in at least one of content of correction and correction amount, for a movement to be corrected, as candidate images of the avatar after being corrected.

2 . The avatar generation system according to claim 1 , wherein

the at least one processor is further configured to execute the instructions to:

detect the state of penetration between the meshes constituting the avatar, based on the data in which the posture of the avatar has been changed according to a defined movement, using a detection model that detects penetration between the meshes constituting the avatar based on the data in which the posture of the avatar has been changed.

3 . The avatar generation system according to claim 2 , wherein

the detection model is a machine learning model that detects the state of penetration between meshes constituting the avatar from an image showing the avatar of which the posture has been changed.

4 . The avatar generation system according to claim 1 , wherein

the at least one processor is further configured to execute the instructions to:

output a result of the detection and the correction amounts.

5 . The avatar generation system according to claim 4 , wherein

the at least one processor is further configured to execute the instructions to:

output an image of the avatar before being corrected and an image of the avatar after being corrected at a timing of the penetration between the meshes.

6 . The avatar generation system according to claim 1 , wherein

the at least one processor is further configured to execute the instructions to:

output an image in which at least one of a corrected portion in the avatar after being corrected and a movement to be corrected is emphasized.

7 . The avatar generation system according to claim 1 , wherein

the at least one processor is further configured to execute the instructions to:

output a plurality of images in which the avatar is moving based on selected movements.

8 . The avatar generation system according to claim 1 , wherein

the at least one processor is further configured to execute the instructions to:

output each image of the avatar in which the same type of mesh is changed.

9 . The avatar generation system according to claim 1 , wherein

the at least one processor is further configured to execute the instructions to:

estimate a correction amount of a setting value of each of a plurality of meshes based on priorities set for at least some of the plurality of meshes.

10 . The avatar generation system according to claim 1 , wherein

the at least one processor is further configured to execute the instructions to:

estimate a correction amount of a setting value of each of a plurality of meshes based on a weight value set for each of the plurality of meshes.

11 . The avatar generation system according to claim 1 , wherein

the state of penetration is at least one of a portion of the penetration between the meshes, an amount of the penetration, and a timing at which the penetration occurs.

12 . The avatar generation system according to claim 1 , wherein

the setting value of each of the meshes is at least one of a shape of the mesh, a position of a vertex of the mesh, and a setting value of a weight of the mesh.

13 . An avatar generation method comprising:

acquiring setting data for bones and meshes of an avatar;

executing processing of changing a posture of the avatar based on the setting data and a defined movement;

detecting a state of penetration between the meshes constituting the avatar based on data in which the posture of the avatar has been changed;

estimating correction amounts of setting values of the meshes based on the result of the detection; and

outputting a plurality of images of the avatar different from each other in at least one of content of correction and correction amount, for a movement to be corrected, as candidate images of the avatar after being corrected.

14 . The avatar generation method according to claim 13 , wherein

the state of penetration between the meshes constituting the avatar is detected based on the data in which the posture of the avatar has been changed according to a defined movement, using a detection model that detects penetration between the meshes constituting the avatar based on the data in which the posture of the avatar has been changed.

15 . The avatar generation method according to claim 13 , further comprising:

outputting a result of the detection and the correction amounts.

16 . The avatar generation method according to claim 15 , wherein

an image of the avatar before being corrected and an image of the avatar after being corrected are output at a timing of the penetration between the meshes.

17 . The avatar generation method according to claim 15 , wherein

an image in which at least one of a corrected portion in the avatar after being corrected and a movement to be corrected is emphasized is output.

18 . A non-transitory recording medium that stores avatar generation program for causing a computer to execute:

acquiring setting data for bones and meshes of an avatar;

executing processing of changing a posture of the avatar based on the setting data and a defined movement;

detecting a state of penetration between the meshes constituting the avatar based on data in which the posture of the avatar has been changed;

estimating correction amounts of setting values of the meshes based on the result of the detection; and

outputting a plurality of images of the avatar different from each other in at least one of content of correction and correction amount, for a movement to be corrected, as candidate images of the avatar after being corrected.