Seminars/Colloquia/Invited Talks

Seminars

Chengjun Liu

Face Detection and Recognition

PLACE:	Clark 314
EVENT:	CIS Seminar Series
DATE:	June 21, 2006
TIME:	9:30 - 10:30 AM

Abstract

In this talk I will discuss the Bayesian Discriminating Features (BDF) method for face detection, and the Dimensionality-Increasing pattern recognition Framework (DIF) for face recognition. The BDF method, which is trained on images from only one database yet works on test images from diverse sources, displays robust generalization performance. The three major components of the BDF method are the discriminating feature analysis of the input image, the statistical modeling of the face and the nonface classes, and the Bayes classifier design. The DIF framework capitalizes on dimensionality increasing techniques by integrating the Gabor image representation, a multi-class Kernel Fisher Analysis (KFA) method, and fractional power polynomial models for improving pattern recognition performance. The DIF framework has been shown effective for solving complex pattern recognition problems, such as the Face Recognition Grand Challenge (FRGC) problems.

Brief biography

Chengjun Liu received the Ph.D. from George Mason University in 1999, and he is presently an Associate Professor of Computer Science at New Jersey Institute of Technology. His recent research focuses on biometrics with an emphasis on face recognition. Some representative methods he has developed are the Bayesian Discriminating Features (BDF) method for face detection, the Gabor-Fisher Classifier (GFC), the Dimensionality-Increasing pattern recognition Framework (DIF) for face recognition, and the Evolutionary Pursuit (EP) method for pattern recognition in general, and face recognition in particular.