Nonlinear features for improved pattern recognition

Nonlinear features that represent higher-order correlations in input data are considered for improved recognition. They optimize new performance measures that do not make Gaussian etc. data distribution assumptions and that are intended for improved discrimination. The new features are produced in closed-form and are thus preferable to iterative solutions. An efficient two-step feature extraction algorithm is presented for the high-dimensional (iconic) input data case of most interest. The feature generation can be realized as a new neural network with adaptive activation functions. Test results on pose-invariant face recognition are emphasized; results on standard feature inputs for a product inspection application are briefly noted as a low- dimensional input data case.

[1]  R. Lippmann,et al.  An introduction to computing with neural nets , 1987, IEEE ASSP Magazine.

[2]  David Casasent,et al.  Nonlinear features for classification and pose estimation of machined parts from single views , 1998, Other Conferences.

[3]  Kazuyuki Hattori,et al.  Estimating pose of human face based on symmetry plane using range and intensity images , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[4]  Bahram Javidi,et al.  Neural networks-based face recognition using Fourier plane nonlinear filters , 1995, Optics & Photonics.

[5]  Trevor Darrell,et al.  Active face tracking and pose estimation in an interactive room , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[6]  Juyang Weng,et al.  Using Discriminant Eigenfeatures for Image Retrieval , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Yoshihiko Hamamoto,et al.  On a theoretical comparison between the orthonormal discriminant vector method and discriminant analysis , 1993, Pattern Recognit..

[8]  R. Fisher,et al.  Contributions to Mathematical Statistics , 1951 .

[9]  John W. Sammon,et al.  An Optimal Set of Discriminant Vectors , 1975, IEEE Transactions on Computers.

[10]  Juha Karhunen,et al.  Representation and separation of signals using nonlinear PCA type learning , 1994, Neural Networks.

[11]  Alex Pentland,et al.  Bayesian face recognition using deformable intensity surfaces , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[12]  Qian Chen,et al.  3D head pose estimation without feature tracking , 1998, Proceedings Third IEEE International Conference on Automatic Face and Gesture Recognition.

[13]  David Casasent,et al.  Nonlinear features for product inspection , 1999, Defense, Security, and Sensing.

[14]  Tomaso A. Poggio,et al.  Linear Object Classes and Image Synthesis From a Single Example Image , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Michael J. Jones,et al.  Gaze estimation using morphable models , 1998, Proceedings Third IEEE International Conference on Automatic Face and Gesture Recognition.