Facial Expression Recognition by Supervised Independent Component Analysis Using MAP Estimation

Permutation ambiguity of the classical Independent Component Analysis (ICA) may cause problems in feature extraction for pattern classification. Especially when only a small subset of components is derived from data, these components may not be most distinctive for classification, because ICA is an unsupervised method. We include a selective prior for de-mixing coefficients into the classical ICA to alleviate the problem. Since the prior is constructed upon the classification information from the training data, we refer to the proposed ICA model with a selective prior as a supervised ICA (sICA). We formulated the learning rule for sICA by taking a Maximum a Posteriori (MAP) scheme and further derived a fixed point algorithm for learning the de-mixing matrix. We investigate the performance of sICA in facial expression recognition from the aspects of both correct rate of recognition and robustness even with few independent components.

[1]  K. Kotani,et al.  Facial Expression Recognition by Fast Supervised Independent Component Analysis , 2007 .

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

[3]  Maja Pantic,et al.  Automatic Analysis of Facial Expressions: The State of the Art , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Christopher J. James,et al.  Electromagnetic brain signal analysis using constrained ICA , 2002 .

[5]  Terrence J. Sejnowski,et al.  The “independent components” of natural scenes are edge filters , 1997, Vision Research.

[6]  Takeo Kanade,et al.  Comprehensive database for facial expression analysis , 2000, Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition (Cat. No. PR00580).

[7]  Frank Y. Shih,et al.  Recognizing facial action units using independent component analysis and support vector machine , 2006, Pattern Recognit..

[8]  Bruce A. Draper,et al.  Recognizing faces with PCA and ICA , 2003, Comput. Vis. Image Underst..

[9]  Shun-ichi Amari,et al.  Natural Gradient Works Efficiently in Learning , 1998, Neural Computation.

[10]  Fan Chen,et al.  Facial Expression Recognition by ICA with Selective Prior , 2006, ICA.

[11]  Mislav Grgic,et al.  Independent comparative study of PCA, ICA, and LDA on the FERET data set , 2005, Int. J. Imaging Syst. Technol..

[12]  Marian Stewart Bartlett,et al.  Image Representations for Facial Expression Coding , 1999, NIPS.

[13]  Manabu Kotani,et al.  Supervised Independent Component Analysis with Class Information , 2004, ICONIP.

[14]  Michael J. Lyons,et al.  Coding facial expressions with Gabor wavelets , 1998, Proceedings Third IEEE International Conference on Automatic Face and Gesture Recognition.

[15]  Bülent Sankur,et al.  Feature selection in the independent component subspace for face recognition , 2004, Pattern Recognit. Lett..

[16]  Kazunori Kotani,et al.  Facial expression analysis by generalized eigen-space method based on class-features (GEMC) , 2005, IEEE International Conference on Image Processing 2005.

[17]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[18]  Beat Fasel,et al.  Automati Fa ial Expression Analysis: A Survey , 1999 .

[19]  Josef Kittler,et al.  Independent component analysis in a local facial residue space for face recognition , 2004, Pattern Recognit..

[20]  P. Ekman,et al.  Emotion in the Human Face: Guidelines for Research and an Integration of Findings , 1972 .

[21]  M. Lewicki,et al.  Learning higher-order structures in natural images , 2003, Network.