Presentation Attack
Presentation attack detection (PAD) aims to secure biometric systems by identifying fraudulent attempts to impersonate a genuine user using fake biometric data (e.g., printed photos, masks, deepfakes). Current research heavily focuses on improving the generalization of PAD models across diverse attack types and unseen domains, employing techniques like convolutional neural networks (CNNs), vision transformers (ViTs), generative adversarial networks (GANs), and various feature fusion methods. The development of robust and generalizable PAD systems is crucial for ensuring the security and reliability of biometric authentication in various applications, from access control to financial transactions.
44papers
Papers
December 10, 2024
September 13, 2024
September 10, 2024
August 23, 2024
April 19, 2024
January 17, 2024
November 23, 2023
November 7, 2023
September 21, 2023
September 1, 2023
August 29, 2023