Session variability modelling for face authentication

• Published in: IET biometrics Vol. 2; no. 3; pp. 117 - 129
• Main Authors: McCool, Christopher, Wallace, Roy, McLaren, Mitchell, El Shafey, Laurent, Marcel, Sébastien
• Format: Journal Article
• Language: English
• Published: Stevenage The Institution of Engineering and Technology 01.09.2013; IET; John Wiley & Sons, Inc
• Subjects: Adaptation; Algorithms; Applied sciences; Authentication; BANCA; Biometrics; Cryptography; detrimental within-class variation; error rate reductions; Exact sciences and technology; Face; face authentication; face recognition; Factor analysis; Gaussian; Gaussian mixture models; Gaussian processes; Information, signal and communications theory; intersession variability modelling; ISV; JFA; joint factor analysis; Mathematical models; Miscellaneous; MOBIO; Modelling; multiPIE; Probabilistic models; publicly available databases; SCface; Signal and communications theory; Signal processing; speaker authentication; speaker recognition; Speech processing; Telecommunications and information theory; Variability; visual databases; JFA; face authentication; MOBIO; visual databases; multiPIE; ISV; BANCA; SCface; detrimental within-class variation; joint factor analysis; error rate reductions; Gaussian processes; speaker authentication; face recognition; Gaussian mixture models; publicly available databases; intersession variability modelling; speaker recognition; Factor analysis; Gaussian process; Audio signal processing; Authentication; Error rate; Database; Mixture theory; Error correction; Speaker recognition; Acoustic signal processing; Modeling; Speech processing
• ISSN: 2047-4938; 2047-4946; 2047-4946
• DOI: 10.1049/iet-bmt.2012.0059

This study examines session variability modelling for face authentication using Gaussian mixture models. Session variability modelling aims to explicitly model and suppress detrimental within-class (inter-session) variation. The authors examine two techniques to do this, inter-session variability modelling (ISV) and joint factor analysis (JFA), which were initially developed for speaker authentication. We present a self-contained description of these two techniques and demonstrate that they can be successfully applied to face authentication. In particular, they show that using ISV leads to significant error rate reductions of, on average, 26% on the challenging and publicly available databases SCface, BANCA, MOBIO and multi-PIE. Finally, the authors show that a limitation of both ISV and JFA for face authentication is that the session variability model captures and suppresses a significant portion of between-class variation.