Adversarial Image Translation: Unrestricted Adversarial Examples in Face Recognition Systems

Authors: Kazuya Kakizaki, Kosuke Yoshida
Published: 2019-05-09 | Updated: 2020-01-28

Source: https://arxiv.org/abs/1905.03421

PDF: https://arxiv.org/pdf/1905.03421

AIにより推定されたラベル

敵対的サンプル 敵対的攻撃検出 性能低下の要因

※ こちらのラベルはAIによって自動的に追加されました。そのため、正確でないことがあります。
詳細は文献データベースについてをご覧ください。

Abstract

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 success rates under white- and black-box settings, respectively, and that the translated images are perceptually realistic and maintain the identifiability of the individual while the perturbations are large enough to bypass certified defenses.