arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2201.10937cs.CV
收藏

利用频率攻击增强3D对抗攻击

Boosting 3D Adversarial Attacks with Attacking On Frequency

Tsinghua University(清华大学);Guangxi University(广西大学)

Binbin Liu, Jinlai Zhang, Lyujie Chen, Jihong Zhu

摘要:

深度神经网络(DNN)已被证明易受对抗攻击的攻击。最近,3D对抗攻击,特别是点云上的对抗攻击,引起了越来越多的兴趣。然而,由以前的方法获得的对抗性点云显示出弱的可传递性,并且易于防御。为了解决这些问题,在本文中,我们提出了一种新的点云攻击(被称为AOF),更加关注点云的低频分量。我们结合点云的损失及其低频分量来制作对抗样本。大量的实验证明,AOF可以提高可移植性相比,国家的最先进的(SOTA)的攻击,并更强大的SOTA 3D防御方法。另外,与干净点云相比,AOF得到的对抗点云包含更多的变形。

英文摘要:

Deep neural networks (DNNs) have been shown to be vulnerable to adversarial attacks. Recently, 3D adversarial attacks, especially adversarial attacks on point clouds, have elicited mounting interest. However, adversarial point clouds obtained by previous methods show weak transferability and are easy to defend. To address these problems, in this paper we propose a novel point cloud attack (dubbed AOF) that pays more attention on the low-frequency component of point clouds. We combine the losses from point cloud and its low-frequency component to craft adversarial samples. Extensive experiments validate that AOF can improve the transferability significantly compared to state-of-the-art (SOTA) attacks, and is more robust to SOTA 3D defense methods. Otherwise, compared to clean point clouds, adversarial point clouds obtained by AOF contain more deformation than outlier.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑