For enterprise, personal and societal applications, there is now an increasing demand for automated authentication of identity from images using computer vision. However, current authentication technologies are still vulnerable to presentation attacks. We present RoPAD, an end-to-end deep learning model for presentation attack detection that employs unsupervised adversarial invariance to ignore visual distractors in images for increased robustness and reduced overfitting. Experiments show that the proposed framework exhibits state-of-the-art performance on presentation attack detection on several benchmark datasets.

RoPAD: Robust Presentation Attack Detection through Unsupervised Adversarial Invariance / Jaiswal, A; Xia, S; Masi, I; Abdalmageed, W. - (2019). (Intervento presentato al convegno IAPR International Conference on Biometrics tenutosi a Crete. Greece).

RoPAD: Robust Presentation Attack Detection through Unsupervised Adversarial Invariance

Masi I;
2019

Abstract

For enterprise, personal and societal applications, there is now an increasing demand for automated authentication of identity from images using computer vision. However, current authentication technologies are still vulnerable to presentation attacks. We present RoPAD, an end-to-end deep learning model for presentation attack detection that employs unsupervised adversarial invariance to ignore visual distractors in images for increased robustness and reduced overfitting. Experiments show that the proposed framework exhibits state-of-the-art performance on presentation attack detection on several benchmark datasets.
2019
IAPR International Conference on Biometrics
face anti spoofing, biometrics, deep learning
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
RoPAD: Robust Presentation Attack Detection through Unsupervised Adversarial Invariance / Jaiswal, A; Xia, S; Masi, I; Abdalmageed, W. - (2019). (Intervento presentato al convegno IAPR International Conference on Biometrics tenutosi a Crete. Greece).
File allegati a questo prodotto
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1458895
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 10
  • ???jsp.display-item.citation.isi??? 0
social impact