Face Recognition and Video Processing (FRVP) lab focuses on the
research and development
of biometric systems, with a special emphasis on face and iris
recognition systems, and video surveillance systems. Some
representative methods we have 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.
The DIF framework, for example, has been shown effective for solving
complex pattern recognition problems, such as the Face Recognition
Grand Challenge (FRGC) problems - a preliminary application of the DIF
framework to face recognition achieves the best performance on the most
challenging FRGC experiment (the 3rd FRGC workshop, February 16, 2005,
McLean, VA).
|
|