MA-ADV:运动感知的点云事件相机对抗攻击框架
Generating Adversarial Events: A Motion-Aware Point Cloud Framework
AI 导读
研究者提出 MA-ADV,首个基于点云表示为事件相机数据生成对抗样本的运动感知框架。该方法用扩散方法平滑扰动、结合逐样本 Adam 优化、迭代精炼与二分搜索找到最小成本扰动,实现 100% 攻击成功率,并对防御具有更强鲁棒性。
正文 · 原文
Abstract:Event cameras have been widely adopted in safety-critical domains such as autonomous driving, robotics, and human-computer interaction. A pressing challenge arises from the vulnerability of deep neural networks to adversarial examples, which poses a significant threat to the reliability of event-based systems. Nevertheless, research into adversarial attacks on events is scarce. This is primarily due to the non-differentiable nature of mainstream event representations, which hinders the extension of gradient-based attack methods. In this paper, we propose MA-ADV, a novel \textbf{M}otion-\textbf{A}ware \textbf{Adv}ersarial framework. To the best of our knowledge, this is the first work to generate adversarial events by leveraging point cloud representations. MA-ADV accounts for high-frequency noise in events and employs a diffusion-based approach to smooth perturbations, while fully leveraging the spatial and temporal relationships among events. Finally, MA-ADV identifies the minimal-cost perturbation through a combination of sample-wise Adam optimization, iterative refinement, and binary search. Extensive experimental results validate that MA-ADV ensures a 100\% attack success rate with minimal perturbation cost, and also demonstrate enhanced robustness against defenses, underscoring the critical security challenges facing future event-based perception systems.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2602.08230 [cs.CV] |
| (or arXiv:2602.08230v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2602.08230 arXiv-issued DOI via DataCite |
Submission history
From: Hongwei Ren [view email]
[v1]
Mon, 9 Feb 2026 03:06:07 UTC (1,586 KB)
[v2]
Wed, 18 Nov 2026 03:40:13 UTC (1,808 KB)
来源:arXiv cs.AI · arxiv.org