Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Iris Presentation Attack Detection by Attention-based and Deep Pixel-wise Binary Supervision Network

: Fang, Meiling; Damer, Naser; Boutros, Fadi; Kirchbuchner, Florian; Kuijper, Arjan


Institute of Electrical and Electronics Engineers -IEEE-; Institute of Electrical and Electronics Engineers -IEEE-, Biometrics Council; International Association for Pattern Recognition -IAPR-:
IEEE International Joint Conference on Biometrics, IJCB 2021 : 4-7 August 2021, Shenzhen, China, virtual
Piscataway, NJ: IEEE, 2021
ISBN: 978-1-6654-3781-3
ISBN: 978-1-6654-3780-6
Art. 9484343, 8 S.
International Joint Conference on Biometrics (IJCB) <2021, Online>
Bundesministerium für Bildung und Forschung BMBF (Deutschland)
Fraunhofer IGD ()
Lead Topic: Digitized Work; Lead Topic: Smart City; Research Line: Computer vision (CV); Research Line: Machine Learning (ML); biometrics; deep learning; machine learning; spoofing attacks; Iris recognition; ATHENE; CRISP

Iris presentation attack detection (PAD) plays a vital role in iris recognition systems. Most existing CNN-based iris PAD solutions 1) perform only binary label supervision during the training of CNNs, serving global information learning but weakening the capture of local discriminative features, 2) prefer the stacked deeper convolutions or expert-designed networks, raising the risk of overfitting, 3) fuse multiple PAD systems or various types of features, increasing difficulty for deployment on mobile devices. Hence, we propose a novel attention-based deep pixel-wise bi-nary supervision (A-PBS) method. Pixel-wise supervision is first able to capture the fine-grained pixel/patch-level cues. Then, the attention mechanism guides the network to automatically find regions that most contribute to an accurate PAD decision. Extensive experiments are performed on LivDet-Iris 2017 and three other publicly available databases to show the effectiveness and robustness of proposed A-PBS methods. For instance, the A-PBS model achieves an HTER of 6.50% on the IIITD-WVU database outperforming state-of-the-art methods.