Share Email Print

Proceedings Paper

Automatic 3D facial model reconstruction from single front-view image
Author(s): Yanchao Xing; Chunxia Guo; Zheng Tan
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

A novel algorithm for automatic 3D facial model reconstruction form a single front-view facial image is presented. First, face and facial features are located. Then feature point depths are estimated from frontal facial measurements through multi-variant linear regression. With these feature points, the generic face model is individualized by volume morphing with a novel Radial Basis Function-based interpolation. Finally, the texture map is generated through cylindrical mapping. Experiments show that the reconstructed model has satisfying visual effects for a broad pose range. Compared with the multi-view based facial model reconstruction, this algorithm is fast, easy to use, fully automatic and has low requirement in hardware, so it has great potential for many practical applications.

Paper Details

Date Published: 19 March 2004
PDF: 4 pages
Proc. SPIE 5444, Fourth International Conference on Virtual Reality and Its Applications in Industry, (19 March 2004); doi: 10.1117/12.561234
Show Author Affiliations
Yanchao Xing, Xi'an Jiaotong Univ. (China)
Chunxia Guo, Xi'an Jiaotong Univ. (China)
Zheng Tan, Xi'an Jiaotong Univ. (China)

Published in SPIE Proceedings Vol. 5444:
Fourth International Conference on Virtual Reality and Its Applications in Industry
Jizhou Sun; Zhigeng Pan, Editor(s)

© SPIE. Terms of Use
Back to Top