CONF norman:icassp:2006/IDIAP Using Chimeric Users to Construct Fusion Classifiers in Biometric Authentication Tasks: An Investigation Poh, Norman Bengio, Samy EXTERNAL https://publications.idiap.ch/attachments/papers/2006/norman-icassp-2006.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/Poh_05_VR_chimeric_fuse_rr Related documents IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP) 2006 IDIAP-RR 05-59 Chimeric users have recently been proposed in the field of biometric person authentication as a way to overcome the problem of lack of real multimodal biometric databases as well as an important privacy issue -- the fact that too many biometric modalities of a same person stored in a single location can present a \emph{higher} risk of identity theft. While the privacy problem is indeed solved using chimeric users, it is still an open question of how such chimeric database can be efficiently used. For instance, the following two questions arise: i) Is the performance measured on a chimeric database a good predictor of that measured on a real-user database?, and, ii) can a chimeric database be exploited to \emph{improve} the generalization performance of a fusion operator on a real-user database?. Based on a considerable amount of empirical biometric person authentication experiments (21 real-user data sets and up to $21 \times 1000$ chimeric data sets and two fusion operators,',','), our previous study~\cite{Poh_05_chimeric} answers {\bf no} to the first question. The current study aims to answer the second question. Having tested on four classifiers and as many as 3380 face and speech bimodal fusion tasks (over 4 different protocols) on the BANCA database and four different fusion operators, this study shows that generating multiple chimeric databases \emph{does not degrade nor improve} the performance of a fusion operator when tested on a real-user database with respect to using only a real-user database. Considering the possibly expensive cost involved in collecting the real-user multimodal data, our proposed approach is thus \emph{useful} to construct a trainable fusion classifier while at the same time being able to overcome the problem of small size training data. REPORT Poh_05_VR_chimeric_fuse_rr/IDIAP Using Chimeric Users to Construct Fusion Classifiers in Biometric Authentication Tasks: An Investigation Poh, Norman Bengio, Samy EXTERNAL https://publications.idiap.ch/attachments/reports/2005/rr05-59.pdf PUBLIC Idiap-RR-59-2005 2005 IDIAP Chimeric users have recently been proposed in the field of biometric person authentication as a way to overcome the problem of lack of real multimodal biometric databases as well as an important privacy issue -- the fact that too many biometric modalities of a same person stored in a single location can present a higher risk of identity theft. While the privacy problem is indeed solved using chimeric users, it is still an open question of how such chimeric database can be efficiently used. For instance, the following two questions arise: i) Is the performance measured on a chimeric database a good predictor of that measured on a real-user database?, and, ii) can a chimeric database be exploited to improve the generalization performance of a fusion operator on a real-user database?. Based on a considerable amount of empirical biometric person authentication experiments (21 real-user data sets and up to 21 times 1000 chimeric data sets and two fusion operators,',','), our previous study answers no to the first question. The current study aims to answer the second question. Having tested on four classifiers and as many as 3380 face and speech bimodal fusion tasks (over 4 different protocols) on the BANCA database and four different fusion operators, this study shows that generating multiple chimeric databases does not degrade nor improve the performance of a fusion operator when tested on a real-user database with respect to using only a real-user database. Considering the possibly expensive cost involved in collecting the real-user multimodal data, our proposed approach is thus useful to construct a trainable fusion classifier while at the same time being able to overcome the problem of small size training data.