The study area of this thesis is face recognition, one of the important fields in computer vision. Although face recognition has recently achieved many advances, the process is still not able to meet the accuracy requirements of many applications that are affected by variations in pose and illumination. The aim of this thesis is to develop a more advanced approach that can handle the challenges in pose and illumination in face recognition. The thesis proposes Robust Multi-Scale Block Local Binary Pattern as a new facial representation that is sufficiently robust to accept variations in pose and illumination and yet contains rich discriminative information. The thesis also investigates the metrics or scores in general used to measure similarity/dissimilarity in face recognition and contributes two novel classification methods, namely Extended Bayesian Learning and Relation Learning, to overcome difficulties such as the Small-Sample-Size problem and gain good performance for face recognition systems.
|Date of Award||2015|
|Supervisor||Dat Tran (Supervisor) & Xu Huang (Supervisor)|