RoPAD: Robust Presentation Attack Detection through Unsupervised Adversarial Invariance
This addresses the problem of robust identity authentication for enterprise, personal, and societal applications, representing an incremental improvement over existing methods.
The paper tackles the vulnerability of current authentication technologies to presentation attacks by introducing RoPAD, an end-to-end deep learning model that uses unsupervised adversarial invariance to ignore visual distractors, resulting in state-of-the-art performance on multiple benchmark datasets.
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.