Granted Patent B2

US 12,412,352 · App. 18/025,551 · Granted Sep 9, 2025

Inventors: Fabien Danieau (Rennes, FR); Glenn Kerbiriou (Cesson-Sevigne, FR); Quentin Avril (Betton, FR); Nicolas Olivier (Chateaubourg, FR)

Assignee: InterDigital CE Patent Holdings, SAS

G06T19/20G06T7/11G06T7/64G06T2200/24G06T2207/20021G06T2207/20096G06T2207/20221G06T2207/30201G06T2210/44G06T2219/2021

Abstract

Claims (46)

1. A method for generating a three-dimensional (3D) face comprising at least one deformed region, the method comprising:

obtaining a curvature exaggeration face from at least one region of a first 3D face;

obtaining a set of relative deformations from a first set of 3D caricatured faces and a second set of 3D faces, wherein the set of relative deformations are the 3D caricatured faces of the first set standardized using a standard deviation and a mean of the 3D faces of the second set;

obtaining a first relative deformation for the at least one region of the first 3D face, wherein the first relative deformation is obtained by a standardization of the at least one region of the first 3D face using the standard deviation and the mean of the 3D faces of the second set;

obtaining, based on the first relative deformation, at least one nearest relative deformation from the set of relative deformations;

determining a proportion exaggeration deformation from the at least one nearest relative deformation, wherein the proportion exaggeration deformation is a weighted average deformation from the at least one nearest relative deformation, based on a distance of the at least one nearest relative deformation to the first relative deformation; and

combining the curvature exaggeration face and the proportion exaggeration deformation for obtaining the at least one deformed region of the 3D face.

2. The method of claim 1 , further comprising:

obtaining at least one user input setting at least one weighting factor for weighting at least one of the curvature exaggeration or the proportion exaggeration deformation.

3. The method of claim 1 , wherein the least one region of the first 3D face corresponds to the first 3D face.

4. The method of claim 1 , wherein the first 3D face comprises a head, and wherein the head comprises at least one eye, a nose and a mouth, the method further comprising:

segmenting the first 3D face into a set of face regions representative of at least the at least one eye, the nose, the mouth and a rest of the head of the first 3D face; and

merging deformed region obtained for the face regions to obtain a deformed 3D face.

5. The method of claim 4 , wherein the set of face regions further comprises the first 3D face.

6. The method of claim 1 , further comprising accessing the first 3D face.

7. The method of claim 1 , wherein the curvature exaggeration face is based on a gradient obtained from the second 3D face.

8. The method of claim 1 , wherein the curvature exaggeration face is reconstructed from the least one region of the first 3D face and a second 3D face distinct from the first 3D face.

9. An apparatus for generating a three-dimensional (3D) face comprising at least one deformed region, the apparatus comprising one or more processors configured to:

obtain a curvature exaggeration face from at least one region of a first 3D face;

obtain a set of relative deformations from a first set of 3D caricatured faces and a second set of 3D faces, wherein the set of relative deformations are the 3D caricatured faces of the first set standardized using a standard deviation and a mean of the 3D faces of the second set;

obtain a first relative deformation for the at least one region of the first 3D face, wherein the first relative deformation is obtained by a standardization of the at least one region of the first 3D face using the standard deviation and the mean of the 3D faces of the second set;

obtain, based on the first relative deformation, at least one nearest relative deformation from the set of relative deformations;

determine a proportion exaggeration deformation from the at least one nearest relative deformation, wherein the proportion exaggeration deformation is a weighted average deformation from the at least one nearest relative deformation, based on a distance of the at least one nearest relative deformation to the first relative deformation; and

combine the curvature exaggeration face and the proportion exaggeration deformation for obtaining the at least one deformed region of the 3D face.

10. The apparatus of claim 9 , wherein the one or more processors are further configured to access the first 3D face.

11. The apparatus of claim 9 , wherein the one or more processors are further configured to obtain at least one user input setting at least one weighting factor for weighting at least one of the curvature exaggeration or the proportion exaggeration deformation.

12. The apparatus of claim 9 , wherein the least one region of the first 3D face corresponds to the first 3D face.

13. The apparatus of claim 9 , wherein the curvature exaggeration face is reconstructed from the least one region of the first 3D face and a second 3D face distinct from the first 3D face.

14. The apparatus of claim 9 , wherein the curvature exaggeration face is based on a gradient obtained from the second 3D face.

15. A system for generating a three-dimensional (3D) face comprising at least one deformed region, the system comprising:

an apparatus according to claim 9 ;

a first data set of 3D faces; and

a second data set of 3D deformed faces.

16. A non-transitory computer readable storage medium having stored thereon instructions for causing one or more processors to perform a method for generating a three-dimensional (3D) face comprising at least one deformed region, the method comprising:

obtaining a curvature exaggeration face from at least one region of a first 3D face;

obtaining a set of relative deformations from a first set of 3D caricatured faces and a second set of 3D faces, wherein the set of relative deformations are the 3D caricatured faces of the first set standardized using a standard deviation and a mean of the 3D faces of the second set;

obtaining a first relative deformation for the at least one region of the first 3D face, wherein the first relative deformation is obtained by a standardization of the at least one region of the first 3D face using the standard deviation and the mean of the 3D faces of the second set;

obtaining, based on the first relative deformation, at least one nearest relative deformation from the set of relative deformations;

determining a proportion exaggeration deformation from the at least one nearest relative deformation, wherein the proportion exaggeration deformation is a weighted average deformation from the at least one nearest relative deformation, based on a distance of the at least one nearest relative deformation to the first relative deformation; and

combining the curvature exaggeration face and the proportion exaggeration deformation for obtaining the at least one deformed region of the 3D face.

17. The non-transitory computer readable storage medium of claim 16 , wherein the curvature exaggeration face is reconstructed from the least one region of the first 3D face and a second 3D face distinct from the first 3D face.

18. The non-transitory computer readable storage medium of claim 16 , wherein the instructions further cause the one or more processors to perform:

accessing the first 3D face.

19. The non-transitory computer readable storage medium of claim 16 , wherein the instructions further cause the one or more processors to perform:

obtaining at least one user input setting at least one weighting factor for weighting at least one of the curvature exaggeration or the proportion exaggeration deformation.

20. The non-transitory computer readable storage medium of claim 16 , wherein the least one region of the first 3D face corresponds to the first 3D face.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (27)