Abstract
Video-based object detection plays a vital role in safety-critical applications. While deep learning-based object detectors have achieved impressive performance, they remain vulnerable to adversarial attacks, particularly those involving universal perturbations. In this work, we propose a minimally distorted universal adversarial attack tailored for video object detection, which leverages nuclear norm regularization to promote structured perturbations concentrated in the background. To optimize this formulation efficiently, we employ an adaptive, optimistic exponentiated gradient method that enhances both scalability and convergence. Our results demonstrate that the proposed attack outperforms both low-rank projected gradient descent and Frank-Wolfe based attacks in effectiveness while maintaining high stealthiness. All code and data are publicly available at this https URL.
Abstract (translated)
基于视频的对象检测在安全关键应用中扮演着重要角色。虽然基于深度学习的对象探测器已经取得了令人印象深刻的性能，但它们仍然容易受到对抗性攻击的影响，特别是那些涉及普遍扰动的攻击。在这项工作中，我们提出了一种针对视频对象检测的、最小化结构变形的通用对抗攻击方法，该方法利用核范数正则化来促进集中在背景中的结构化扰动。为了高效地优化这一公式，我们采用了一种自适应的乐观指数梯度法，这种方法增强了可扩展性和收敛性。我们的研究结果表明，在保持高度隐蔽性的前提下，所提出的攻击方法在有效性上超过了低秩投影梯度下降和基于Frank-Wolfe的方法。所有代码和数据都可以通过以下链接公开获取：[此URL]。 请确保替换上述提示中的“this https URL”为实际的链接地址以供访问。
URL
https://arxiv.org/abs/2510.14460
