New framework enhances physical adversarial attacks on vision models

Researchers have developed a new framework called JMOF to create more effective physical adversarial attacks against computer vision models. This framework addresses the issue of attacks overfitting to single models by using a joint multi-objective and multi-model optimization approach. It also incorporates an Orthogonal Gradient Alignment strategy to resolve conflicts between different models' gradients, enhancing the attack's ability to generalize across various vision tasks like object detection and semantic segmentation. AI

影响 This research advances the generalization limits of physical adversarial attacks, providing a robust framework for evaluating visual AI vulnerabilities in real-world deployments.

排序理由 The cluster contains an academic paper detailing a new framework for adversarial attacks. [lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

• Computer Vision

• 论文
• 安全

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →