Relativistic Discriminator: A One-Class Classifier For Generalized Iris Presentation Attack Detection
2020 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV)(2020)
摘要
Iris based recognition systems are vulnerable to presentation attacks (PAs) where artifacts such as cosmetic contact lenses, artificial eyes and printed eyes can be used to fool the system. While many learning-based algorithms have been proposed to detect such attacks, very few are equipped to handle previously unseen or newly constructed PAs. In this research, we propose a presentation attack detection (PAD) method that utilizes a discriminator that is trained to distinguish between bonafide iris images and synthetically generated iris images. We hypothesize that such a discriminator will generate a tight boundary around the bonafide samples. This would allow the discriminator to better separate the bonafide samples from all types of PA samples. For generating synthetic irides, we train the Relativistic Average Standard Generative Adversarial Network (RaSGAN) that has been shown to generate higher resolution and better quality images than standard GANs. The relativistic discriminator (RD) component of the trained RaSGAN is then appropriated for PA detection and is referred to as RD-PAD. Experimental results convey the efficacy of the RD-PAD as a one-class anomaly detector.
更多查看译文
关键词
relativistic discriminator,PA detection,RD-PAD,one-class classifier,generalized iris presentation attack detection,learning-based algorithms,Relativistic Average Standard Generative Adversarial Network,image quality,RaS-GAN,iris based recognition systems
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络