Realistic 3D facial animation parameters from mirror-reflected multi-view video

Author

Lin, I-Chen ; Yeh, Jeng-Sheng ; Ouhyoung, Ming

Author_Institution

Dept. of Comput. Sci. & Inf. Eng., Nat. Taiwan Univ., Taipei, Taiwan

fYear

2001

fDate

2001

Firstpage

2

Lastpage

250

Abstract

A robust, accurate and inexpensive approach to estimate 3D facial motion from multi-view video is proposed, where two mirrors located near one´s cheeks can reflect the side views of markers on one face. Nice properties of mirrored images are utilized to simplify the proposed tracking algorithm significantly, while a Kalman filter is employed to reduce the noise and to predict the occluded marker positions. More than 50 markers on one face are continuously tracked at 30 frames per second. The estimated 3D facial motion data has been practically applied to our facial animation system. In addition, the dataset of facial motion can also be applied to the analysis of co-articulation effects, facial expressions, and audio-visual hybrid recognition system

Keywords

Kalman filters; computer animation; image sequences; motion estimation; realistic images; 3D facial motion estimation; Kalman filter; audio-visual hybrid recognition system; co-articulation effects; estimated 3D facial motion data; facial animation system; facial expressions; facial motion dataset; mirror-reflected multi-view video; mirrored images; multi-view video; occluded marker positions; realistic 3D facial animation parameters; tracking algorithm; Acoustic scattering; Face; Facial animation; Head; Humans; Mirrors; Motion estimation; Noise reduction; Robustness; Video sequences;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Animation, 2001. The Fourteenth Conference on Computer Animation. Proceedings

Conference_Location

Seoul

ISSN

1087-4844

Print_ISBN

0-7803-7237-9

Type

conf

DOI

10.1109/CA.2001.982371

Filename

982371