Fully Automatic Facial Deformation Transfer

Greg Maguire (Contributor), Shaojun Bian (Contributor), Anzong Zheng (Contributor), Lin Gao (Contributor), Willem Kokke (Contributor), Jon Macey (Contributor), Lihua You (Contributor), Jianjun Zhang (Contributor)

Research output: Contribution to journalArticle

Abstract

Facial Animation is a serious and ongoing challenge for the Computer Graphic industry. Because diverse and complex emotions need to be expressed by different facial deformation and animation, copying facial deformations from existing character to another is widely needed in both industry and academia, to reduce time-consuming and repetitive manual work of modeling to create the 3D shape sequences for every new character. But transfer of realistic facial animations between two 3D models is limited and inconvenient for general use. Modern deformation transfer methods require correspondences mapping, in most cases, which are tedious to get. In this paper, we present a fast and automatic approach to transfer the deformations of the facial mesh models by obtaining the 3D point-wise correspondences in the automatic manner. The key idea is that we could estimate the correspondences with different facial meshes using the robust facial landmark detection method by projecting the 3D model to the 2D image. Experiments show that without any manual labelling efforts, our method detects reliable correspondences faster and simpler compared with the state-of-the-art automatic deformation transfer method on the facial models.
LanguageEnglish
Pages27
Number of pages10
JournalSymmetry
Volume12
Issue number1
Publication statusPublished - 21 Dec 2019

Fingerprint

Animation
Copying
Computer graphics
Labeling
Industry
Experiments

Keywords

  • face landmark detection
  • orthographic projection
  • point-wise correspondences
  • automatic deformation transfer

Cite this

Maguire, G., Bian, S., Zheng, A., Gao, L., Kokke, W., Macey, J., ... Zhang, J. (2019). Fully Automatic Facial Deformation Transfer. Symmetry, 12(1), 27.
Maguire, Greg ; Bian, Shaojun ; Zheng, Anzong ; Gao, Lin ; Kokke, Willem ; Macey, Jon ; You, Lihua ; Zhang, Jianjun. / Fully Automatic Facial Deformation Transfer. In: Symmetry. 2019 ; Vol. 12, No. 1. pp. 27.
@article{52d7236f8e2b4061b2b3940e523dba29,
title = "Fully Automatic Facial Deformation Transfer",
abstract = "Facial Animation is a serious and ongoing challenge for the Computer Graphic industry. Because diverse and complex emotions need to be expressed by different facial deformation and animation, copying facial deformations from existing character to another is widely needed in both industry and academia, to reduce time-consuming and repetitive manual work of modeling to create the 3D shape sequences for every new character. But transfer of realistic facial animations between two 3D models is limited and inconvenient for general use. Modern deformation transfer methods require correspondences mapping, in most cases, which are tedious to get. In this paper, we present a fast and automatic approach to transfer the deformations of the facial mesh models by obtaining the 3D point-wise correspondences in the automatic manner. The key idea is that we could estimate the correspondences with different facial meshes using the robust facial landmark detection method by projecting the 3D model to the 2D image. Experiments show that without any manual labelling efforts, our method detects reliable correspondences faster and simpler compared with the state-of-the-art automatic deformation transfer method on the facial models.",
keywords = "face landmark detection, orthographic projection, point-wise correspondences, automatic deformation transfer",
author = "Greg Maguire and Shaojun Bian and Anzong Zheng and Lin Gao and Willem Kokke and Jon Macey and Lihua You and Jianjun Zhang",
year = "2019",
month = "12",
day = "21",
language = "English",
volume = "12",
pages = "27",
journal = "Symmetry",
issn = "2073-8994",
publisher = "MDPI",
number = "1",

}

Maguire, G, Bian, S, Zheng, A, Gao, L, Kokke, W, Macey, J, You, L & Zhang, J 2019, 'Fully Automatic Facial Deformation Transfer', Symmetry, vol. 12, no. 1, pp. 27.

Fully Automatic Facial Deformation Transfer. / Maguire, Greg (Contributor); Bian, Shaojun (Contributor); Zheng, Anzong (Contributor); Gao, Lin (Contributor); Kokke, Willem (Contributor); Macey, Jon (Contributor); You, Lihua (Contributor); Zhang, Jianjun (Contributor).

In: Symmetry, Vol. 12, No. 1, 21.12.2019, p. 27.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fully Automatic Facial Deformation Transfer

A2 - Maguire, Greg

A2 - Bian, Shaojun

A2 - Zheng, Anzong

A2 - Gao, Lin

A2 - Kokke, Willem

A2 - Macey, Jon

A2 - You, Lihua

A2 - Zhang, Jianjun

PY - 2019/12/21

Y1 - 2019/12/21

N2 - Facial Animation is a serious and ongoing challenge for the Computer Graphic industry. Because diverse and complex emotions need to be expressed by different facial deformation and animation, copying facial deformations from existing character to another is widely needed in both industry and academia, to reduce time-consuming and repetitive manual work of modeling to create the 3D shape sequences for every new character. But transfer of realistic facial animations between two 3D models is limited and inconvenient for general use. Modern deformation transfer methods require correspondences mapping, in most cases, which are tedious to get. In this paper, we present a fast and automatic approach to transfer the deformations of the facial mesh models by obtaining the 3D point-wise correspondences in the automatic manner. The key idea is that we could estimate the correspondences with different facial meshes using the robust facial landmark detection method by projecting the 3D model to the 2D image. Experiments show that without any manual labelling efforts, our method detects reliable correspondences faster and simpler compared with the state-of-the-art automatic deformation transfer method on the facial models.

AB - Facial Animation is a serious and ongoing challenge for the Computer Graphic industry. Because diverse and complex emotions need to be expressed by different facial deformation and animation, copying facial deformations from existing character to another is widely needed in both industry and academia, to reduce time-consuming and repetitive manual work of modeling to create the 3D shape sequences for every new character. But transfer of realistic facial animations between two 3D models is limited and inconvenient for general use. Modern deformation transfer methods require correspondences mapping, in most cases, which are tedious to get. In this paper, we present a fast and automatic approach to transfer the deformations of the facial mesh models by obtaining the 3D point-wise correspondences in the automatic manner. The key idea is that we could estimate the correspondences with different facial meshes using the robust facial landmark detection method by projecting the 3D model to the 2D image. Experiments show that without any manual labelling efforts, our method detects reliable correspondences faster and simpler compared with the state-of-the-art automatic deformation transfer method on the facial models.

KW - face landmark detection

KW - orthographic projection

KW - point-wise correspondences

KW - automatic deformation transfer

M3 - Article

VL - 12

SP - 27

JO - Symmetry

T2 - Symmetry

JF - Symmetry

SN - 2073-8994

IS - 1

ER -

Maguire G, Bian S, Zheng A, Gao L, Kokke W, Macey J et al. Fully Automatic Facial Deformation Transfer. Symmetry. 2019 Dec 21;12(1):27.