RoPAD: Robust Presentation Attack Detection through Unsupervised Adversarial Invariance

03/08/2019
by   Ayush Jaiswal, et al.
0

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.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset