物理对抗服装通过非重叠 RGB-T 图案规避可见光-热红外检测器
Physical Adversarial Clothing Evades Visible-Thermal Detectors via Non-Overlapping RGB-T Pattern
对做 RGB-T 多模态检测的人，这篇给出了一个必须正视的物理世界威胁模型：用可见光与热红外材料不重叠的服装图案即可在 0°–360° 视角下稳定欺骗不同融合架构的检测器，并提示融合阶段集成可提升跨架构迁移性，可作为评估多模态融合鲁棒性的红队基线。
原文依据 we introduce a fusion-stage ensemble method that enhances the transferability of adversarial attacks across unseen RGB-T detectors
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论可见光-热红外（RGB-T）目标检测是自动驾驶等应用中的关键技术，多模态融合可在低光照等困难条件下提升性能。然而，RGB-T 检测器的安全性，尤其是物理世界中的安全性，长期被忽视。本文提出一种利用具有非重叠 RGB-T 图案（NORP）的对抗服装实施 RGB-T 物理攻击的新方法。为模拟全视角（0°–360°）RGB-T 攻击，我们构建了人体与对抗服装的三维 RGB-T 模型。NORP 是一种新的对抗图案设计，使用互不重叠的可见光材料与热红外材料，从而避免重叠式 RGB-T 图案（ORP）带来的光强衰减。为在对抗服装上优化 NORP，我们提出空间离散-连续优化（SDCO）方法。我们在具有不同融合架构的 RGB-T 检测器上系统评估了该方法，在数字世界与物理世界中均取得高攻击成功率。此外，我们引入一种融合阶段集成方法，提升了对抗攻击在具有不同融合架构的未见 RGB-T 检测器上的迁移性。
英文原文摘要 1206 字符
Visible-thermal (RGB-T) object detection is a crucial technology for applications such as autonomous driving, where multimodal fusion enhances performance in challenging conditions like low light. However, the security of RGB-T detectors, particularly in the physical world, has been largely overlooked. This paper proposes a novel approach to RGB-T physical attacks using adversarial clothing with a non-overlapping RGB-T pattern (NORP). To simulate full-view (0^ \circ -360^ \circ ) RGB-T attacks, we construct 3D RGB-T models for human and adversarial clothing. NORP is a new adversarial pattern design using distinct visible and thermal materials without overlap, avoiding the light reduction in overlapping RGB-T patterns (ORP). To optimize the NORP on adversarial clothing, we propose a spatial discrete-continuous optimization (SDCO) method. We systematically evaluated our method on RGB-T detectors with different fusion architectures, demonstrating high attack success rates both in the digital and physical worlds. Additionally, we introduce a fusion-stage ensemble method that enhances the transferability of adversarial attacks across unseen RGB-T detectors with different fusion architectures.