Estimation of Face Depths by Conditional Densities

The expected value of missing data in a sample taken from a multivariate normal probability distribution is the mean of the conditi onal distribution of the missing dimensions given the known dimensions. We explain the derivation of this result, demonstrate its application to face image p rocessing, then use it in a new method for recovering shape from image data. The context of our work is the use of 3D facial models to aid in recognition of human faces by humans. We explain the requirement for such models and review the prac tical possibilities for encoding depth information alongside photographs in identity documents like passports. The best alternative is to derive depths automatically from the photos, as this requires no side information. We show exper imentally that conditional density estimation provides accurate face depth rec overy, without recourse to explicit modelling of surface shape.

[1]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[2]  Richard A. Johnson,et al.  Applied Multivariate Statistical Analysis , 1983 .

[3]  David J. Kriegman,et al.  Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection , 1996, ECCV.

[4]  Lawrence Sirovich,et al.  The global dimensionality of face space , 2000, Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition (Cat. No. PR00580).

[5]  C. Liu,et al.  Face recognition with perspective transformation , 2003, Vision Research.

[6]  D Kersten,et al.  Viewpoint-Dependent Recognition of Familiar Faces , 1999, Perception.

[7]  Frances L. Krouse.  Effects of pose, pose change, and delay on face recognition performance. , 1981 .

[8]  V. Bruce,et al.  Effects of lighting on the perception of facial surfaces. , 1996, Journal of experimental psychology. Human perception and performance.

[9]  P. N. Bellhumer.  Eigenfaces vs. fisherfaces : Recognition using class specific linear projection , 1997 .

[10]  J. Friedman.  Regularized Discriminant Analysis , 1989 .

[11]  C. Liu,et al.  Does Face Recognition Rely on Encoding of 3-D Surface? Examining the Role of Shape-from-Shading and Shape-from-Stereo , 2000, Perception.

[12]  V. Bruce,et al.  The basis of the 3/4 view advantage in face recognition , 1987 .

[13]  Bor-Chen Kuo,et al.  A covariance estimator for small sample size classification problems and its application to feature extraction , 2002, IEEE Trans. Geosci. Remote. Sens..

[14]  G. Pike,et al.  When Seeing should not be Believing: Photographs, Credit Cards and Fraud , 1997 .

[15]  John Robinson,et al.  Coding 3D Facial Models for Mugshot Applications , 2003, VVG.

[16]  Alex Pentland,et al.  Probabilistic Visual Learning for Object Representation , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  V. Bruce,et al.  Effects of lighting on the perception of facial surfaces. , 1996, Journal of experimental psychology. Human perception and performance.

[18]  E. Tulving.  Elements of episodic memory , 1983 .

[19]  V. Bruce,et al.  Recognition of unfamiliar faces , 2000, Trends in Cognitive Sciences.

[20]  Lawrence Sirovich,et al.  Application of the Karhunen-Loeve Procedure for the Characterization of Human Faces , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  H. Bülthoff,et al.  Face recognition under varying poses: The role of texture and shape , 1996, Vision Research.