Mirror PCA: Exploiting Facial Symmetry for Feature Extraction

Feature extraction technique aiming at obtaining discriminative information from high-dimensional face images is of great importance in face recognition. One widely used method for extracting primary feature is Principal Component Analysis (PCA), which uses projection matrix for dimensionality reduction. There are many improvements of PCA but no one pays attention to the fact that both facial images and facial expression are symmetrical to some degree. Facial symmetry is a helpful characteristic, which benefits of feature extraction. In this paper, Mirror Principal Component Analysis (Mirror PCA) method is proposed for extracting representative facial features, which takes advantage of the facial symmetry in a face image. In order to verify the effectiveness of the proposed method, we compare the Mirror PCA method with other four methods on four famous face databases. The experimental results indicate that the representation capacity of our method is superior to others.

[1]  P. Ekman,et al.  The symmetry of emotional and deliberate facial actions. , 1981, Psychophysiology.

[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]  Hong Liu,et al.  Using the original and 'symmetrical face' training samples to perform representation based two-step face recognition , 2013, Pattern Recognit..

[4]  Alejandro F. Frangi,et al.  Two-dimensional PCA: a new approach to appearance-based face representation and recognition , 2004 .

[5]  Chung-Lin Huang,et al.  A content-based image retrieval system , 1998, Image Vis. Comput..

[6]  Lei Wang,et al.  Generalized 2D principal component analysis for face image representation and recognition , 2005, Neural Networks.

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

[8]  I. Jolliffe.  Principal Component Analysis , 2002 .

[9]  Xuelong Li,et al.  Robust Tensor Analysis With L1-Norm , 2010, IEEE Transactions on Circuits and Systems for Video Technology.

[10]  David Zhang,et al.  Represent and fuse bimodal biometric images at the feature level: complex-matrix-based fusion scheme , 2010 .

[11]  Chris H. Q. Ding,et al.  R1-PCA: rotational invariant L1-norm principal component analysis for robust subspace factorization , 2006, ICML.

[12]  Xiaoqing Ding,et al.  Symmetrical PCA in face recognition , 2002, Proceedings. International Conference on Image Processing.

[13]  David J. Kriegman,et al.  Acquiring linear subspaces for face recognition under variable lighting , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Wai Keung Wong,et al.  Approximate Low-Rank Projection Learning for Feature Extraction , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[15]  Nojun Kwak,et al.  Principal Component Analysis Based on L1-Norm Maximization , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Harry Wechsler,et al.  The FERET database and evaluation procedure for face-recognition algorithms , 1998, Image Vis. Comput..

[17]  A. Ng.  Feature selection, L1 vs. L2 regularization, and rotational invariance , 2004, Twenty-first international conference on Machine learning - ICML '04.

[18]  Jinwen Ma,et al.  ICA and PCA integrated feature extraction for classification , 2016, 2016 IEEE 13th International Conference on Signal Processing (ICSP).

[19]  Sung-Jea Ko,et al.  High-dimensional feature extraction using bit-plane decomposition of local binary patterns for robust face recognition , 2017, J. Vis. Commun. Image Represent..

[20]  Eli Saber,et al.  Frontal-view face detection and facial feature extraction using color, shape and symmetry based cost functions , 1998, Pattern Recognit. Lett..

[21]  Haixian Wang,et al.  Block principal component analysis with L1-norm for image analysis , 2012, Pattern Recognit. Lett..

[22]  Takeo Kanade,et al.  Robust L/sub 1/ norm factorization in the presence of outliers and missing data by alternative convex programming , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[23]  Adel M. Alimi,et al.  Human identity recognition using sparse auto encoder for texture information representation in palmprint images based on voting technique , 2017, 2017 Sudan Conference on Computer Science and Information Technology (SCCSIT).

[24]  Feiping Nie,et al.  Robust Principal Component Analysis with Non-Greedy l1-Norm Maximization , 2011, IJCAI.

[25]  Liming Chen,et al.  Discriminative Transfer Learning Using Similarities and Dissimilarities , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[26]  Xuelong Li,et al.  Low-Rank Preserving Projections , 2016, IEEE Transactions on Cybernetics.

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

[28]  Xuelong Li,et al.  L1-Norm-Based 2DPCA , 2010, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[29]  Young Gil Kim,et al.  Face recognition using a fusion method based on bidirectional 2DPCA , 2008, Appl. Math. Comput..

[30]  Chong-Ho Choi,et al.  Image covariance-based subspace method for face recognition , 2007, Pattern Recognit..