Discriminant non-negative graph embedding for face recognition

Abstract Non-negative Matrix Factorization (NMF) is an unsupervised algorithm for low-rank approximation of non-negative data and has been widely used in many fields, but its performance in feature extraction is not satisfactory. The main reason is that the model of NMF and its variants did not take into account the label information of the samples, which can add the discriminant ability of the methods. In this paper, we proposed a novel method, called discriminant non-negative graph embedding (DNGE) algorithm in which the label information of the samples and the local geometric structure are all integrated in the objective function. Furthermore, we incorporated the between-class graph and within-class graph into the objective functions to indicate that we not only used the local separability but also used the whole separability of the samples. To guarantee convergence, we use the KKT condition to calculate the non-negative solution of the DNGE. A convergent multiplicative non-negative updating rule is then derived to learn the transformation matrix. Experiments are conducted on the CMU PIE, ORL, Yale, FERET and AR database. The results show that the DNGE algorithm provides better facial representation and achieves higher recognition rates than naive Non-Negative Matrix Factorization and its extension methods.

[1]  Wei Liu,et al.  Nonnegative Local Coordinate Factorization for Image Representation , 2011, IEEE Transactions on Image Processing.

[2]  Seungjin Choi,et al.  Manifold-respecting discriminant nonnegative matrix factorization , 2011, Pattern Recognit. Lett..

[3]  Patrik O. Hoyer,et al.  Non-negative Matrix Factorization with Sparseness Constraints , 2004, J. Mach. Learn. Res..

[4]  Anastasios Tefas,et al.  Exploiting discriminant information in nonnegative matrix factorization with application to frontal face verification , 2006, IEEE Transactions on Neural Networks.

[5]  Fengxi Song,et al.  A novel local preserving projection scheme for use with face recognition , 2010, Expert Syst. Appl..

[6]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

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

[8]  Shiliang Sun,et al.  Tangent space intrinsic manifold regularization for data representation , 2013, 2013 IEEE China Summit and International Conference on Signal and Information Processing.

[9]  I. Pitas,et al.  Discriminant NMFfaces for Frontal Face Verification , 2005, 2005 IEEE Workshop on Machine Learning for Signal Processing.

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

[11]  Kun Zhou,et al.  Locality Sensitive Discriminant Analysis , 2007, IJCAI.

[12]  Hujun Bao,et al.  Non-negative local coordinate factorization for image representation , 2011, CVPR.

[13]  Jiawei Han,et al.  Non-negative Matrix Factorization on Manifold , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[14]  Hai Jin,et al.  Projective Nonnegative Graph Embedding , 2010, IEEE Transactions on Image Processing.

[15]  Ivor W. Tsang,et al.  Flexible Manifold Embedding: A Framework for Semi-Supervised and Unsupervised Dimension Reduction , 2010, IEEE Transactions on Image Processing.

[16]  Erkki Oja,et al.  Projective Nonnegative Matrix Factorization for Image Compression and Feature Extraction , 2005, SCIA.

[17]  D. Kalman.  A Singularly Valuable Decomposition: The SVD of a Matrix , 1996 .

[18]  Jiawei Han,et al.  Learning a Maximum Margin Subspace for Image Retrieval , 2008, IEEE Transactions on Knowledge and Data Engineering.

[19]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[20]  Anastasios Tefas,et al.  Using subclasses in discriminant non-negative subspace learning for facial expression recognition , 2011, 2011 19th European Signal Processing Conference.

[21]  Mikhail Belkin,et al.  Laplacian Eigenmaps and Spectral Techniques for Embedding and Clustering , 2001, NIPS.

[22]  Shiliang Sun,et al.  Manifold-preserving graph reduction for sparse semi-supervised learning , 2014, Neurocomputing.

[23]  Aleix M. Martinez,et al.  The AR face database , 1998 .

[24]  Thomas S. Huang,et al.  Graph Regularized Nonnegative Matrix Factorization for Data Representation. , 2011, IEEE transactions on pattern analysis and machine intelligence.

[25]  Seungjin Choi,et al.  Semi-Supervised Nonnegative Matrix Factorization , 2010, IEEE Signal Processing Letters.

[26]  Xiaofei He,et al.  Locality Preserving Projections , 2003, NIPS.

[27]  Markus Flierl,et al.  Graph-Preserving Sparse Nonnegative Matrix Factorization With Application to Facial Expression Recognition , 2011, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[28]  Yong Xu,et al.  One improvement to two-dimensional locality preserving projection method for use with face recognition , 2009, Neurocomputing.

[29]  Yunde Jia,et al.  Non-negative matrix factorization framework for face recognition , 2005, Int. J. Pattern Recognit. Artif. Intell..

[30]  J. Tenenbaum,et al.  A global geometric framework for nonlinear dimensionality reduction. , 2000, Science.

[31]  Jian Yang,et al.  LPP solution schemes for use with face recognition , 2010, Pattern Recognit..

[32]  Laiyuan Xiao,et al.  Supervised Discriminant Nonnegative Matrix Factorization Method , 2009, 2009 Second International Symposium on Knowledge Acquisition and Modeling.