Kazuya Kakizaki creator text xx 2019 monographic ng

print

Thanks to recent advances in deep neural networks (DNNs), face recognition systems have become highly accurate in classifying a large number of face images. However, recent studies have found that DNNs could be vulnerable to adversarial examples, raising concerns about the robustness of such systems. Adversarial examples that are not restricted to small perturbations could be more serious since conventional certified defenses might be ineffective against them. To shed light on the vulnerability to such adversarial examples, we propose a flexible and efficient method for generating unrestricted adversarial examples using image translation techniques. Our method enables us to translate a source image into any desired facial appearance with large perturbations to deceive target face recognition systems. Our experimental results indicate that our method achieved about $90$ and $80\%$ attack Open access — freely available to read. https://arxiv.org/pdf/1905.03421v3 https://arxiv.org/pdf/1905.03421v3 Open access — freely available to read. 250101 20260901030334.0