3D Facial Shape Reconstruction from a Single Frontal Image

Article Preview

Abstract:

In this paper, we propose an improved method for reconstruct 3D facial shape from a single frontal image. We use improved fast marching method to solve the Eikonal equation which can obtained from the method of shape from shading (SFS). In order to overcome the concave-convex ambiguity problems inherent to SFS, we find out the concave region and recover concave into the convex keeping the relative shape invariant to reconstruct the accuracy facial shape.

You might also be interested in these eBooks

Info:

Periodical:

Advanced Materials Research (Volumes 989-994)

Pages:

3544-3547

Citation:

Online since:

July 2014

Authors:

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2014 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] Horn B. K. P. Obtaining shape from shading information. Chapter 4 in The Psychology of Computer Vision, P.H. Winston(ed. ), McGraw Hill, New York, 1975, pp.115-155.

Google Scholar

[2] G. A. Sor, V. K. Dhir, Numerical Method in Fluid Dynamic, Cambridge University Press, (1985).

Google Scholar

[3] PRADOS E, FAUGERAS O. Shape from shading: a well-posed problem? [C]. IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos, CA, USA: IEEE Computer Society, 2005, pp.870-877.

DOI: 10.1109/cvpr.2005.319

Google Scholar

[4] O. J. Liensis, Uniqueness in shape from shading, International Journal of Computer Vision. 1991, 6 (2), pp.75-104.

Google Scholar

[5] H. Gouraud, Computer Display of Curved Surfaees, IEEE Conf. on Computer Graphics, Pattern Recognition, and Data Structure, Los Angeles, California, 1975, 5, pp.11-17.

Google Scholar

[6] Osher S., Fedkiw R. Level Set Methods and Dynamic Implicit Surfaces[M]. Springer-Verlag, (2002).

Google Scholar