Search papers, labs, and topics across Lattice.
This paper introduces a novel physical adversarial attack against RGB-T object detectors using adversarial clothing with a non-overlapping RGB-T pattern (NORP) to avoid light reduction issues of overlapping patterns. They propose a spatial discrete-continuous optimization (SDCO) method to optimize the NORP on 3D models of clothing and humans. Experiments demonstrate high attack success rates in both digital and physical settings, along with a fusion-stage ensemble method to improve transferability across different RGB-T detector architectures.
Adversarial clothing with non-overlapping visible-thermal patterns can reliably evade RGB-T detectors, even transferring across different fusion architectures.
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.
BAIR
Tsinghua AI