论文标题
对面部识别的对抗性攻击：一项全面研究
Adversarial Attacks against Face Recognition: A Comprehensive Study
论文作者
论文摘要
面部识别（FR）系统已显示出出色的验证性能，这表明对现实世界应用的适用性从社交媒体中的照片标记到自动边界控制（ABC）。但是，在具有深度学习架构的先进FR系统中，仅促进识别效率就不够，并且该系统还应承受旨在针对其熟练程度的潜在攻击。最近的研究表明，（深）FR系统表现出令人着迷的脆弱性，这些脆弱性无法察觉或可感知但外观自然的对抗性输入图像，这些图像将模型推向了不正确的输出预测。在本文中，我们对针对FR系统的对抗性攻击进行了全面调查，并详细介绍了针对他们的新对策能力。此外，我们提出了基于不同标准的现有攻击和防御方法的分类法。我们将攻击方法在类别的方向和属性和防御方法上进行比较。最后，我们探讨了挑战和潜在的研究方向。
Face recognition (FR) systems have demonstrated outstanding verification performance, suggesting suitability for real-world applications ranging from photo tagging in social media to automated border control (ABC). In an advanced FR system with deep learning-based architecture, however, promoting the recognition efficiency alone is not sufficient, and the system should also withstand potential kinds of attacks designed to target its proficiency. Recent studies show that (deep) FR systems exhibit an intriguing vulnerability to imperceptible or perceptible but natural-looking adversarial input images that drive the model to incorrect output predictions. In this article, we present a comprehensive survey on adversarial attacks against FR systems and elaborate on the competence of new countermeasures against them. Further, we propose a taxonomy of existing attack and defense methods based on different criteria. We compare attack methods on the orientation and attributes and defense approaches on the category. Finally, we explore the challenges and potential research direction.
