Discriminative Learning in Biometrics

This monograph describes the latest advances in discriminative learning methods for biometric recognition.

Specifically, it focuses on three representative categories of methods: sparse representation-based classification, metric learning, and discriminative feature representation, together with their applications in palmprint authentication, face recognition and multi-biometrics.

The ideas, algorithms, experimental evaluation and underlying rationales are also provided for a better understanding of these methods.

Lastly, it discusses several promising research directions in the field of discriminative biometric recognition.