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 2412.07511cs.CV
收藏

通过附加点特征对三维点云进行隐蔽且鲁棒的后门攻击

Stealthy and Robust Backdoor Attack against 3D Point Clouds through Additional Point Features

Wuhan University of Technology(武汉理工大学);Engineering Research Center of Intelligent Service Technology for Digital Publishing, Ministry of Education(教育部数字出版智能服务技术工程研究中心);University of Electronic Science and Technology of China(电子科技大学);Wuhan Vocational College of Software and Engineering(武汉软件工程职业学院)

Xiaoyang Ning, Qing Xie, Jinyu Xu, Wenbo Jiang, Jiachen Li, Yanchun Ma

AI总结:

针对三维点云后门攻击易被防御和察觉的问题,提出SRBA方法,利用附加点特征统一偏移作为触发器,在不改变几何信息下实现隐蔽性,并通过贝叶斯优化自动化攻击,攻击成功率超94%。

摘要:

近年来,三维后门攻击对面向三维点云设计的三维深度神经网络(3D DNNs)构成了重大威胁,这些网络广泛部署于各类安全关键型应用中。尽管现有的三维后门攻击取得了较高的攻击性能,但它们仍然容易受到基于预处理的防御手段(如离群点移除和旋转增强)的影响,并且容易被人工检查所察觉。为了追求一种更难防御且更隐蔽的三维后门攻击,本文提出了隐蔽且鲁棒的后门攻击(SRBA),通过有意的设计考量来确保鲁棒性和隐蔽性。我们攻击的关键思路是对点云的附加点特征(如反射强度)施加统一偏移,这些特征被广泛用作三维深度神经网络输入的一部分,并以此作为触发器。在不改变点云几何信息的情况下,我们的攻击确保了中毒样本与良性样本之间的视觉一致性,并展示了对基于预处理的防御手段的鲁棒性。此外,为了实现攻击的自动化,我们采用贝叶斯优化(BO)来识别合适的触发器。大量实验表明,SRBA在所有情况下均实现了超过94%的攻击成功率(ASR),并且在训练过程中应用多种预处理操作时,显著优于先前的最先进(SOTA)方法。

英文摘要:

Recently, 3D backdoor attacks have posed a substantial threat to 3D Deep Neural Networks (3D DNNs) designed for 3D point clouds, which are extensively deployed in various security-critical applications. Although the existing 3D backdoor attacks achieved high attack performance, they remain vulnerable to preprocessing-based defenses (e.g., outlier removal and rotation augmentation) and are prone to detection by human inspection. In pursuit of a more challenging-to-defend and stealthy 3D backdoor attack, this paper introduces the Stealthy and Robust Backdoor Attack (SRBA), which ensures robustness and stealthiness through intentional design considerations. The key insight of our attack involves applying a uniform shift to the additional point features of point clouds (e.g., reflection intensity) widely utilized as part of inputs for 3D DNNs as the trigger. Without altering the geometric information of the point clouds, our attack ensures visual consistency between poisoned and benign samples, and demonstrate robustness against preprocessing-based defenses. In addition, to automate our attack, we employ Bayesian Optimization (BO) to identify the suitable trigger. Extensive experiments suggest that SRBA achieves an attack success rate (ASR) exceeding 94% in all cases, and significantly outperforms previous SOTA methods when multiple preprocessing operations are applied during training.

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。

↑