A novel model for two-dimensional independent TV-Based Quotient image and its application to face recognition

A novel model for Two-Dimensional Independent TV- Based Quotient Image is proposed in this paper, and the new proposed model is applied to face recognition with only one sample per subject on a large scale face database. In our model, an ETVQI model is firstly proposed to overcome drawbacks of the TVQI model. Then, a two-dimensional independent subspace analysis algorithm is proposed to reduce the dimension of face samples normalized by ETVQI model and to extract high-order statistical features of these samples. According to experiments on the CAS-PEAL face database, our proposed model could outperform ICA with architecture II (ICA2), (2D)2PCA and TVQI. And it also confirms that our model is robust not only to illumination, but to other outliers (expression, masking, occlusion etc.) in face recognition.

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

[2]  Daoqiang Zhang,et al.  (2D)2PCA: Two-directional two-dimensional PCA for efficient face representation and recognition , 2005, Neurocomputing.

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

[4]  Dorin Comaniciu,et al.  Illumination normalization for face recognition and uneven background correction using total variation based image models , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

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

[6]  Wen Gao,et al.  The CAS-PEAL Large-Scale Chinese Face Database and Baseline Evaluations , 2008, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[7]  Jian Yang,et al.  Kernel ICA: An alternative formulation and its application to face recognition , 2005, Pattern Recognit..

[8]  Marian Stewart Bartlett,et al.  Face recognition by independent component analysis , 2002, IEEE Trans. Neural Networks.

[9]  Tony F. Chan,et al.  Aspects of Total Variation Regularized L[sup 1] Function Approximation , 2005, SIAM J. Appl. Math..

[10]  Terrence J. Sejnowski,et al.  Blind separation and blind deconvolution: an information-theoretic approach , 1995, 1995 International Conference on Acoustics, Speech, and Signal Processing.

[11]  Wen Gao,et al.  CAS-PEAL Large-Scale Chinese Face Database and Evaluation Protocols , 2004 .

[12]  Dorin Comaniciu,et al.  Total variation models for variable lighting face recognition , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.