Faculty of Engineering and Natural Sciences
3D Facial Feature Localization for Registration
Albert Ali Salah
Accurate automatical localization of fiducial points in face images is an important step in registration. Although statistical methods of landmark localization reach high accuracies with 2D face images, their performances rapidly deteriorate under illumination changes. 3D information can assist this process by either removing the illumination effects from the 2D image, or by supplying robust features based on depth or curvature. We inspect both approaches for this problem. Our results indicate that using 3D features is more promising than illumination correction with the help of 3D. We complement our statistical feature detection scheme with a structural correction scheme and report our results on the FRGC face dataset.
Albert Ali Salah has graduated from Istanbul Alman Lisesi in 1994, obtained his BS (1998) and MS (2000) degrees from Boğaziçi University, Computer Engineering Department. He is a PhD student under the supervision of Prof. Lale Akarun and a research assistant at the Perceptual Intelligence Laboratory of the same department. His PhD research focuses on a biologically motivated approach to 2D-3D face recognition, which relates to computer vision, image processing, machine learning, biometrics, morphometrics, and cognitive science.
July 27, 2006, 10:40, FENS 2019