Abstract
in recent years, fingerprint-based biometric systems have grown rapidly as they are used for various applications such as mobile payments, international border security, and financial transactions. Although the widespread of these systems, it has been found that they are vulnerable to presentation attacks (i.e., spoof attacks). Therefore, improving the generalization ability of fingerprint PAD over unknown materials and unknown sensors is of primary importance. In this work, we proposed a fingerprint PAD with improved cross-sensor and cross-material generalization based on state-of-the-art CNN network; i.e., EfficientNet combined with Generative Adversarial Network (GANs). We will validate the proposed methodologies on the public LivDet2015 dataset provided by the liveness detection competition.