Advanced Correlation Filters for Robust Face Recognition

Research Assistant Professor
Electrical & Computer Eng and CyLab
Carnegie Mellon University

Host: Anil K. Jain

Advanced Correlation Filters have had great success in Automatic Target Recognition (ATR) applications. They have evolved dramatically from the simple matched filter to build in specific type of distortion tolerance via well defined optimization criteria. In this talk I will briefly review the evolution of different Advanced Correlation Filters and how they have been successfully applied to Biometric Identification. Specifically I will focus on successful design and implementation of different filters for face recognition. Many face recognition algorithms suffer under unseen illumination conditions and occlusions whereas different Advanced Correlation Filters such as the Minimum Average Correlation Energy (MACE) filter has built-in tolerance to such distortions and exhibits graceful degradation to occlusions. I will also explain some of the recent work performed in FRGC where a new efficient dimensionality feature extraction scheme has been developed using Advanced Correlation Filters and how Kernel methods have been blended into these methods to provide enhanced discrimination. Other example hybrid methods such as Corefaces- A shift invariant Hybrid PCA correlation filter will also be reviewed for scene searching and detection.

Professor Marios Savvides is currently a Research Assistant Professor at the Electrical and Computer Engineering (ECE) Department at Carnegie Mellon University with a joint appointment at Carnegie Mellon CyLab. Professor Savvides obtained his B.Eng (Hons) degree in Microelectronics Systems Engineering from University of Manchester Institute of Science and Technology (UMIST) in England in 1997, his M.Sc degree in Robotics in 2000 from the Robotics Institute in Carnegie Mellon University (CMU), Pittsburgh and his Ph.D. in May 2004 from the Electrical and Computer Engineering Department, also at CMU, where he focused on Biometrics Identification technology for his thesis "Reduced Complexity Face Recognition using Advanced Correlation Filters and Fourier Subspace Methods for Biometric Applications".

Prior to his current position he was a Systems Scientist in ECE and CyLab at CMU from May 2004 until January 2006. He was also the technical lead in CMU's face recognition and iris efforts in FRGC+FRVT2006 and ICE1.0+ICE2006. His research interests are in Biometric recognition of Face, Iris, Fingerprint and Palmprint modalities. He is a member of IEEE and SPIE and serves on the Program Committee of several Biometric conferences including IEEE AutoID, SPIE Defense & Security Biometric Identification Technologies and International Conference on Image Analysis and Recognition (ICIAR) and the Biometrics Symposium of the annual Biometrics Consortium (BC 2006). He also served on the Biometrics Panel in SPIE's Defense & Security Symposium 2006 and is also listed in 2005 Edition of Marquis' Who's Who in America. He has authored and co-authored over 75 conference and journal articles in the Biometrics area, has filed two patent applications and authored one book chapter and is in the process of writing two more in the Biometrics field.