Towards exaggerated image stereotypes

Given a training set of images and a binary classifier, we introduce the notion of an exaggerated image stereotype for some image class of interest, which emphasizes/exaggerates the characteristic patterns in an image and visualizes which visual information the classification relies on. This is useful for gaining insight into the classification mechanism. The exaggerated image stereotypes results in a proper trade-off between classification accuracy and likelihood of being generated from the class of interest. This is done by optimizing an objective function which consists of a discriminative term based on the classification result, and a generative term based on the assumption of the class distribution. We use this idea with Fisher's Linear Discriminant rule, and assume a multivariate normal distribution for samples within a class. The proposed framework has been applied on handwritten digit data, illustrating specific features differentiating digits. Then it is applied to a face dataset using Active Appearance Model (AAM), where male faces stereotypes are evolved from initial female faces.

[1]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[2]  Gunnar Rätsch,et al.  Kernel PCA pattern reconstruction via approximate pre-images. , 1998 .

[3]  Sandip Sen,et al.  PLEASE: A Prototype Learning System Using Genetic Algorithms , 1995, ICGA.

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

[5]  Song-Chun Zhu.  Filters, Random Fields and Maximum Entropy (FRAME): Towards a Unified Theory for Texture Modeling , 1998 .

[6]  Gunnar Rätsch,et al.  Kernel PCA and De-Noising in Feature Spaces , 1998, NIPS.

[7]  Chi-Jen Lu,et al.  Adaptive Prototype Learning Algorithms: Theoretical and Experimental Studies , 2006, J. Mach. Learn. Res..

[8]  Mads Nielsen,et al.  Feature-Based Image Analysis , 2003, International Journal of Computer Vision.

[9]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[10]  Kurt Stadlthanner,et al.  KPCA denoising and the pre-image problem revisited , 2008, Digit. Signal Process..

[11]  Mads Nielsen,et al.  What Do Features Tell about Images? , 2001, Scale-Space.

[12]  Lars Kai Hansen,et al.  Input space regularization stabilizes pre-images for kernel PCA de-noising , 2009, 2009 IEEE International Workshop on Machine Learning for Signal Processing.

[13]  Mikkel B. Stegmann,et al.  The IMM Face Database, An Annotated Dataset of 240 Face Images , 2004 .

[14]  Timothy F. Cootes,et al.  Active Appearance Models , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Gunnar Rätsch,et al.  Input space versus feature space in kernel-based methods , 1999, IEEE Trans. Neural Networks.

[16]  David G. Stork,et al.  Pattern Classification , 1973 .