Pattern detection with information-based maximum discrimination and error bootstrapping

We have previously (1996, 1997) introduced a visual learning technique based on information-theoretic entropy. In that approach, positive and negative examples of a class of visual patterns were analyzed to obtain the probability model that best discriminate the class among others. Such models were tested in the context of maximum likelihood detection of faces and facial features. In this paper we further improve on that technique by using other family of probability model and by extending the optimization criteria to allow for error bootstrapping. The results include a detail analysis of the improvements obtained and a comparison of these pattern recognition algorithms.

[1]  Françoise Fogelman-Soulié,et al.  Multi-Modular Neural Network Architectures: Applications in Optical Character and Human Face Recognition , 1993, Int. J. Pattern Recognit. Artif. Intell..

[2]  Federico Girosi,et al.  Training support vector machines: an application to face detection , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[3]  Thomas S. Huang,et al.  Maximum likelihood face detection , 1996, Proceedings of the Second International Conference on Automatic Face and Gesture Recognition.

[4]  Tomaso A. Poggio,et al.  Example-Based Learning for View-Based Human Face Detection , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Takeo Kanade,et al.  Neural Network-Based Face Detection , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Thomas S. Huang,et al.  Object detection using hierarchical MRF and MAP estimation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[7]  Thomas S. Huang,et al.  Face detection with information-based maximum discrimination , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[8]  R. Gray.  Entropy and Information Theory , 1990, Springer New York.