Large-scale evaluation of multimodal biometric authentication using state-of-the-art systems

By R. Snelick, Umut Uludag, A. Mink, M. Indovina, Anil K. Jain. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005

588 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Signal Processing

View Paper: https://www.semanticscholar.org/paper/97ee90e28934b56f2b5e122e57b7b5cba9c823da

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Information Systems

In Signal Processing

Citations in the corpus

Evaluation of biometric spoofing in a multimodal system (2010, methodology): “We use the assumption that face and fingerprint traits are independent to create a ‘virtual multimodal dataset’ by combining two different unimodal datasets containing no users in common, similarly to the process undertaken in [16].”
Likelihood Ratio-Based Biometric Score Fusion (2008, background): “Choice of the normalization scheme and combination weights is data-dependent and requires extensive empirical evaluation [4], [6], [7].”
Classifier ensembles: Select real-world applications (2008, methodology): “, [34] have used both a global model of this type but also user-specific models for speech recognition and combined them with the goal of improving recognition rates [56,33,31,106,32,64,101,84].”
Score normalization in multimodal biometric systems (2005, methodology): “The results of Snelick et al. [ 29 ] also show a similar trend.”
An Online System of Multispectral Palmprint Verification (2010, background): “If we take d′i = widi as the normalized distance for band i, then we can extend our proposed score-level fusion scheme to weighted sum: multiply the original distance with the weight for normalization and then substitute the new distance into (5) or (6).”
Heterogeneous information fusion: A novel fusion paradigm for biometric systems (2011, unknown): “An improved scheme [31] considers the weight associated to each base classifier as inversely proportional to the “d-prime” statistic which characterizes the degree of separability between the genuine user and the impostor classes for a given user.”