Revisiting Doddington's Zoo: A Systematic Method to Assess User-dependent Variabilities. Poh, N., Bengio, S., & Ross, A. In Second Workshop on Multimodal User Authentication, MMUA, 2006.
Revisiting Doddington's Zoo: A Systematic Method to Assess User-dependent Variabilities [link]Paper  abstract   bibtex   
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 \emphhigher 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 \emphimprove 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~i̧tePoh_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 \emphdoes 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 \emphuseful to construct a trainable fusion classifier while at the same time being able to overcome the problem of small size training data.
@inproceedings{poh:2006:mmua,
  author = {N. Poh and S. Bengio and A. Ross},
  title = {Revisiting {D}oddington's Zoo: A Systematic Method to Assess User-dependent Variabilities},
  booktitle = {Second Workshop on Multimodal User Authentication, {MMUA}},
  year = 2006,
  url = {publications/ps/poh_2006_mmua.ps.gz},
  pdf = {publications/pdf/poh_2006_mmua.pdf},
  djvu = {publications/djvu/poh_2006_mmua.djvu},
  original = {2006/zoo_mmua},
  topics = {multimodal,biometric_authentication},
  abstract = {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.},
  categorie = {C},
}

Downloads: 0