Authors:
Mariusz Leszczyński
and
Władysław Skarbek
Affiliation:
Warsaw University of Technology, Poland
Keyword(s):
Face verification, Linear Discriminant Analysis, face metrics, Receiver Operating Characteristics.
Related
Ontology
Subjects/Areas/Topics:
Biometrics and Pattern Recognition
;
Image and Video Processing, Compression and Segmentation
;
Multidimensional Signal Processing
;
Multimedia
;
Multimedia Signal Processing
;
Telecommunications
Abstract:
Impact of light conditions on face verification are considered for three linear discriminant feature extraction schemes. Two verification scenarios, the single image query and multi image query, were compared. The extraction algorithms are based on compositions of feature projections on global, intra and inter-class error subspaces: Linear Discriminant Analysis LDA, Dual Linear Discriminant Analysis DLDA, and their combination LDA+DLDA. The metrics for evaluation of the verification error is the Mahalanobis distance between normalized feature vectors. The normalization of feature vectors is justified with the upper bound by Fisher separation index for feature vectors. Experiments conducted on facial databases with complex background show the high performance of DLDA and DLDA+LDA verifiers with Equal Error Rate EER less than one percent. The degradation of results, when controlled light conditions are replaced by uncontrolled ones, is of factor two.