1 January 2008 Face authentication using a hybrid approach
Author Affiliations +
J. of Electronic Imaging, 17(1), 011003 (2008). doi:10.1117/1.2885149
Abstract
We present a hybrid approach to face feature extraction based on the trace transform and the novel kernel partial leastsquares discriminant analysis (KPA). The hybrid approach, called trace kernel partial least-squares discriminant analysis (TKPA), first uses a set of 15 trace functionals to derive robust and discriminative facial features and then applies the KPA method to reduce their dimensionality. The feasibility of the proposed approach was successfully tested on the XM2VTS database, where a false rejection rate (FRR) of 1.25% and a false acceptance rate (FAR) of 2.11% were achieved in our best-performing face authentication experiment. The experimental results also show that the proposed approach can outperform kernel methods such as generalized discriminant analysis (GDA), kernel Fisher analysis (KFA), and complete kernel Fisher discriminant analysis (CKFA) as well as combinations of these methods with features extracted using the trace transform.
Vitomir Štruc, F. Mihelic, N. Pavesic, "Face authentication using a hybrid approach," Journal of Electronic Imaging 17(1), 011003 (1 January 2008). http://dx.doi.org/10.1117/1.2885149
JOURNAL ARTICLE
11 PAGES


SHARE
KEYWORDS
Feature extraction

Matrices

Facial recognition systems

Databases

Principal component analysis

Analytical research

Biometrics

RELATED CONTENT

Multimedia medical archiving system
Proceedings of SPIE (November 21 1995)
Automatic seamless image mosaic method based on SIFT features
Proceedings of SPIE (February 28 2017)
Color Images Of Inhomogeneous Transparent Media
Proceedings of SPIE (September 14 1977)

Back to Top