PulseAugur
LIVE 13:47:32
research · [2 sources] ·
0
research

Physical Adversarial Clothing Evades Visible-Thermal Detectors via Non-Overlapping RGB-T Pattern

By PulseAugur Editorial · Summary by None from 2 sources

Researchers have developed a novel method for physical adversarial attacks against visible-thermal (RGB-T) object detectors, commonly used in applications like autonomous driving. The approach utilizes specially designed clothing with a non-overlapping RGB-T pattern (NORP) that evades detection in both digital and physical environments. This NORP design, optimized using a spatial discrete-continuous optimization (SDCO) method, aims to improve attack success rates across various fusion architectures. AI

Summary written by None from 2 sources. How we write summaries →

IMPACT This research highlights potential security vulnerabilities in multimodal AI systems, particularly those used in safety-critical applications like autonomous driving.

RANK_REASON This is a research paper detailing a novel method for adversarial attacks on computer vision systems.

Read on arXiv cs.CV →

COVERAGE [2]

  1. arXiv cs.CV TIER_1 · Xiaopei Zhu, Guanning Zeng, Zhanhao Hu, Jun Zhu, Xiaolin Hu ·

    Physical Adversarial Clothing Evades Visible-Thermal Detectors via Non-Overlapping RGB-T Pattern

    arXiv:2605.04675v1 Announce Type: new Abstract: 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 dete…

  2. arXiv cs.CV TIER_1 · Xiaolin Hu ·

    Physical Adversarial Clothing Evades Visible-Thermal Detectors via Non-Overlapping RGB-T Pattern

    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 b…

RELATED ENTITIES

RELATED TOPICS