Face Recognition using Discriminatively Trained Orthogonal Rank One Tensor Projections

We propose a method for face recognition based on a discriminative linear projection. In this formulation images are treated as tensors, rather than the more conventional vector of pixels. Projections are pursued sequentially and take the form of a rank one tensor, i.e., a tensor which is the outer product of a set of vectors. A novel and effective technique is proposed to ensure that the rank one tensor projections are orthogonal to one another. These constraints on the tensor projections provide a strong inductive bias and result in better generalization on small training sets. Our work is related to spectrum methods, which achieve orthogonal rank one projections by pursuing consecutive projections in the complement space of previous projections. Although this may be meaningful for applications such as reconstruction, it is less meaningful for pursuing discriminant projections. Our new scheme iteratively solves an eigenvalue problem with orthogonality constraints on one dimension, and solves unconstrained eigenvalue problems on the other dimensions. Experiments demonstrate that on small and medium sized face recognition datasets, this approach outperforms previous embedding methods. On large face datasets this approach achieves results comparable with the best, often using fewer discriminant projections.

[1]  L. Duchene,et al.  An Optimal Transformation for Discriminant and Principal Component Analysis , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Andy Harter,et al.  Parameterisation of a stochastic model for human face identification , 1994, Proceedings of 1994 IEEE Workshop on Applications of Computer Vision.

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

[4]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

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

[6]  David J. Kriegman,et al.  From Few to Many: Illumination Cone Models for Face Recognition under Variable Lighting and Pose , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  David J. Kriegman,et al.  Special issue on face recognition , 2003, Computer Vision and Image Understanding.

[8]  Terence Sim,et al.  The CMU Pose, Illumination, and Expression Database , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Hwann-Tzong Chen,et al.  Local discriminant embedding and its variants , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[10]  Yuxiao Hu,et al.  Face recognition using Laplacianfaces , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Deng Cai,et al.  Tensor Subspace Analysis , 2005, NIPS.

[12]  Dong Xu,et al.  Coupled kernel-based subspace learning , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[13]  Chiou-Shann Fuh,et al.  Learning effective image metrics from few pairwise examples , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[14]  Jiawei Han,et al.  Orthogonal Laplacianfaces for Face Recognition , 2006, IEEE Transactions on Image Processing.

[15]  Stephen Lin,et al.  Rank-one Projections with Adaptive Margins for Face Recognition , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).