In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
will defend his dissertation proposal
Face Recognition in the Presence of Facial Expressions and Accessories
Face recognition (FR) is a technology in which a computing device either classifies human identity based on a facial image (face identification) or verifies whether two images belong to the same subject (face verification). Over the last two decades, the FR research community has achieved remarkable performance when comparing images that are both frontal and non-occluded. However, significant challenges remain in the presence of variations in pose, expression, and occlusions. The goal of this thesis is to achieve statistically significant improvement in the performance of face recognition systems using 2D images that depict individuals with facial expressions and accessories. The specific objectives are (i) Design, implement, and evaluate an architecture of a 3D-aided face recognition system that is modular, easy to use, and easy to install; (ii) Develop and evaluate an algorithm for landmark detection on 2D images; (iii) Develop and evaluate an algorithm for 3D face reconstruction from a single and multiple 2D facial images that depict individuals with open mouth and facial accessories. In this proposal, we will introduce our work to date towards these objectives. Currently, the extended CBL FR system achieved 9% improvement when compared with the performance of FaceNet. The proposed face reconstruction algorithm for a single image achieved at least 10% improvement over E2FAR, a representative of the current state-of-the-art.
Date: Friday, April 27, 2018
Time: 11:00 AM
Place: HBS 317
Advisors: Dr. Ioannis A. Kakadiaris
Faculty, students, and the general public are invited.