Dissertation Defense - University of Houston
Skip to main content

Dissertation Defense

In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

Xiang Xu

will defend his dissertation

Face Recognition in the Presence of Variance in Pose, Expression, and Occlusions


Abstract

Face recognition is a technology in which a computing device either classifies human identity based on a facial image or verifies whether two images belong to the same subject. The recent advances have 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 dissertation is to achieve statistically significant improvement in the performance of face recognition systems using 2D images that depict individuals with facial expressions and accessories. Four contributions made in this dissertation can be summarized as follows: (i) a 3D-aided 2D face recognition system with an additional evaluation package that is modular, easy to use, and easy to install is designed, implemented, and evaluated. The proposed system can work on the facial images in the pose as large as 90 degrees and improve the face-recognition performance by 9% on average when compared with FaceNet on a UHDB31 dataset. (ii) two landmark detectors are developed and evaluated on 2D images that are fast and accurate; (iii) feature aggregation learning is proposed for face reconstruction from a single image that depicts individuals with variances in pose, expression, and occlusion, which achieved 16% and 10% improvement when compared with the current state-of-the-art BU-3DFE and JNU-3D datasets, respectively; and (iv) an occlusion-aware face-recognition approach is proposed that improves the generalization ability of the facial-embedding generator and a graph neural network is designed in an unsupervised manner to adapt the knowledge learned in the image-based scenario to the mixed-media set scenario. The proposed methods achieve statistically significant improvements when compared with the baselines.


Date: Monday, April 22, 2019
Time: 10:00 AM - 12:00 PM
Place: HBS 317
Advisor: Dr. Ioannis A. Kakadiaris

Faculty, students, and the general public are invited.