TY - GEN
T1 - Can we do better in unimodal biometric systems? A novel rank-based score normalization framework for multi-sample galleries
AU - Moutafis, Panagiotis
AU - Kakadiaris, Ioannis A.
PY - 2013
Y1 - 2013
N2 - The large amount of research on multimodal systems raises an important question: can we extract additional information from unimodal systems? In this paper, we propose a rank-based score normalization framework that addresses this problem when multi-sample galleries are avail-able. The main idea is to partition the matching scores into subsets and normalize each subset independently. In addition, we present two versions of our framework that: (i) use gallery-based information (i.e., gallery versus gallery scores), and (ii) update available information in an online fashion. We use the theory of Stochastic Dominance to illustrate that the proposed framework can increase the system's performance. Our approach: (i) does not require tuning of any parameters, (ii) can be used in conjunction with any score normalization technique and any integration rule, and (iii) extends the use of W-score normalization to multisample galleries. While our approach is better suited for an Open-set Identification task, we also demonstrate that it can be used for a Verification task. In order to assess the performance of the proposed framework we conduct experiments using the BDCP Face database. Our approach improves the Detection and Identification Rate by 14.87% for Z-score and by 4.82%for W-score.
AB - The large amount of research on multimodal systems raises an important question: can we extract additional information from unimodal systems? In this paper, we propose a rank-based score normalization framework that addresses this problem when multi-sample galleries are avail-able. The main idea is to partition the matching scores into subsets and normalize each subset independently. In addition, we present two versions of our framework that: (i) use gallery-based information (i.e., gallery versus gallery scores), and (ii) update available information in an online fashion. We use the theory of Stochastic Dominance to illustrate that the proposed framework can increase the system's performance. Our approach: (i) does not require tuning of any parameters, (ii) can be used in conjunction with any score normalization technique and any integration rule, and (iii) extends the use of W-score normalization to multisample galleries. While our approach is better suited for an Open-set Identification task, we also demonstrate that it can be used for a Verification task. In order to assess the performance of the proposed framework we conduct experiments using the BDCP Face database. Our approach improves the Detection and Identification Rate by 14.87% for Z-score and by 4.82%for W-score.
UR - http://www.scopus.com/inward/record.url?scp=84887449420&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84887449420&partnerID=8YFLogxK
U2 - 10.1109/ICB.2013.6612983
DO - 10.1109/ICB.2013.6612983
M3 - Conference contribution
AN - SCOPUS:84887449420
SN - 9781479903108
T3 - Proceedings - 2013 International Conference on Biometrics, ICB 2013
BT - Proceedings - 2013 International Conference on Biometrics, ICB 2013
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 6th IAPR International Conference on Biometrics, ICB 2013
Y2 - 4 June 2013 through 7 June 2013
ER -