Special Sessions

#1 Adaptive Biometric Systems: Recent Advances and Challenges

While biometric technology continues to be adopted, an intrinsic characteristic of the technology is that system error rate cannot attain absolute zero in real-world applications. The main cause of mismatch errors is the variable acquisition conditions, human-sensor interactions  and physiology. In addition to complex operational environments that change over time, biometric models (i.e., a set of templates or statistical models) are typically designed a priori using limited reference data and knowledge of underlying data distributions. Therefore, these models are often poor representatives of the current biometric trait to be recognized, and should be adapted over time in response to new or changing reference samples and other environmental variables. Several innovative supervised, semi-supervised and unsupervised techniques have recently emerged to adapt the biometric system over time. These techniques are collectively termed as adaptive biometric systems. A key issue with adaptive biometric systems is the possibility of corrupting biometric models with impostor samples due to intrinsic system failure or adversary attacks.  Therefore, this special session is devoted to recent advances in the area of adaptive biometric systems and related technologies.

Topics

Suggested topics include the following areas as they apply to adaptive biometrics, but are not limited to:

• Adaptive Biometric Recognition Approaches, Algorithms and Techniques
• Incremental Learning of Features, Data Samples and Classes
• On-Line, Adaptive and Life-Long Learning
• Multi-Classifier Systems
• Multi-Modal Biometric Fusion
• Dynamic Selection and Fusion of Classifiers and Ensembles
• Recognition in Changing Environments
• Novelty and Change Detection

Special Session Organizers and Co-Chairs:

	Eric Granger École de technologie supérieure, Université du Québec, Montreal, Canada Email:eric.granger@etsmtl.ca
	Ajita Rattani Michigan State University, East Lansing, USA Email:ajita@msu.edu