Multilinear Analysis of Image Ensembles: TensorFaces

Natural images are the composite consequence of multiple factors related to scene structure, illumination, and imaging. Multilinear algebra, the algebra of higher-order tensors, offers a potent mathematical framework for analyzing the multifactor structure of image ensembles and for addressing the difficult problem of disentangling the constituent factors or modes. Our multilinear modeling technique employs a tensor extension of the conventional matrix singular value decomposition (SVD), known as the N-mode SVD. As a concrete example, we consider the multilinear analysis of ensembles of facial images that combine several modes, including different facial geometries (people), expressions, head poses, and lighting conditions. Our resulting "TensorFaces" representation has several advantages over conventional eigenfaces. More generally, multilinear analysis shows promise as a unifying framework for a variety of computer vision problems.

[1]  I. G. Priest.  THE OPTICAL SOCIETY OF AMERICA. , 1940, Science.

[2]  L. Tucker,et al.  Some mathematical notes on three-mode factor analysis , 1966, Psychometrika.

[3]  J. Leeuw,et al.  Principal component analysis of three-mode data by means of alternating least squares algorithms , 1980 .

[4]  H. Neudecker,et al.  An approach ton-mode components analysis , 1986 .

[5]  L Sirovich,et al.  Low-dimensional Procedure for the Characterization of Human Faces , 1986 .

[6]  J. Kruskal.  Rank, decomposition, and uniqueness for 3-way and n -way arrays , 1989 .

[7]  M. Turk,et al.  Eigenfaces for Recognition , 1991, Journal of Cognitive Neuroscience.

[8]  J. Magnus,et al.  Matrix Differential Calculus with Applications in Statistics and Econometrics , 1991 .

[9]  B A Wandell,et al.  Linear models of surface and illuminant spectra. , 1992, Journal of the Optical Society of America. A, Optics and image science.

[10]  Patrick C. Teo,et al.  Perceptual image distortion , 1994, Proceedings of 1st International Conference on Image Processing.

[11]  Alex Pentland,et al.  View-based and modular eigenspaces for face recognition , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[12]  Rama Chellappa,et al.  Human and machine recognition of faces: a survey , 1995, Proc. IEEE.

[13]  S. Ullman,et al.  Generalization to Novel Images in Upright and Inverted Faces , 1993, Perception.

[14]  Joshua B. Tenenbaum,et al.  Separating Style and Content , 1996, NIPS.

[15]  Joshua B. Tenenbaum,et al.  Learning bilinear models for two-factor problems in vision , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[16]  L. Lathauwer,et al.  Signal Processing based on Multilinear Algebra , 1997 .

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

[18]  Christopher M. Bishop,et al.  Mixtures of Probabilistic Principal Component Analyzers , 1999, Neural Computation.

[19]  Joshua B. Tenenbaum,et al.  Separating Style and Content with Bilinear Models , 2000, Neural Computation.

[20]  Joos Vandewalle,et al.  On the Best Rank-1 and Rank-(R1 , R2, ... , RN) Approximation of Higher-Order Tensors , 2000, SIAM J. Matrix Anal. Appl..

[21]  Michael I. Jordan,et al.  Mixtures of Probabilistic Principal Component Analyzers , 2001 .

[22]  Amnon Shashua,et al.  Linear image coding for regression and classification using the tensor-rank principle , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[23]  M. Alex O. Vasilescu.  Human motion signatures for character animation , 2001, SIGGRAPH 2001.

[24]  Tamara G. Kolda,et al.  Orthogonal Tensor Decompositions , 2000, SIAM J. Matrix Anal. Appl..

[25]  M. Alex O. Vasilescu.  Human motion signatures: analysis, synthesis, recognition , 2002, Object recognition supported by user interaction for service robots.