A Reminiscence of “Mastermind”: Iris/Periocular Biometrics by “In-Set” CNN Iterative Analysis

Convolutional neural networks (CNNs) have emerged as the most popular classification models in biometrics research. Under the discriminative paradigm of pattern recognition, CNNs are used typically in one of two ways: 1) <italic>verification</italic> mode (“ <italic>are samples from the same person?</italic> ”), where pairs of images are provided to the network to distinguish between <italic>genuine</italic> and <italic>impostor</italic> instances and 2) <italic>identification</italic> mode (“ <italic>whom is this sample from?</italic> ”), where appropriate feature representations that map images to identities are found. This paper postulates a novel mode for using CNNs in biometric identification, by learning models that answer the question “ <italic>is the query’s identity among this set?</italic> ”. The insight is a reminiscence of the classical <italic>Mastermind</italic> game: by iteratively analyzing the network responses when multiple random samples of <inline-formula> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> gallery elements are compared to the query, we obtain weakly correlated matching scores that, altogether, provide solid cues to infer the most likely identity. In this setting, identification is regarded as a variable selection and regularization problem, with sparse linear regression techniques being used to infer the matching probability with respect to each gallery identity. As main strength, this strategy is highly robust to <italic>outlier</italic> matching scores, which are known to be a primary error source in biometric recognition. Our experiments were carried out in full versions of two well-known irises near-infrared (CASIA-IrisV4-Thousand) and periocular visible wavelength (UBIRIS.v2) datasets, and confirm that recognition performance can be solidly boosted-up by the proposed algorithm, when compared with the traditional working modes of CNNs in biometrics.