Detector Collapse: Backdooring Object Detection to Catastrophic Overload or Blindness

arXiv:2404.11357v1 Announce Type: new
Abstract: Object detection tasks, crucial in safety-critical systems like autonomous driving, focus on pinpointing object locations. These detectors are known to be susceptible to backdoor attacks. However, existing backdoor techniques have primarily been adapted from classification tasks, overlooking deeper vulnerabilities specific to object detection. This paper is dedicated to bridging this gap by introducing Detector Collapse} (DC), a brand-new backdoor attack paradigm tailored for object detection. DC is designed to instantly incapacitate detectors (i.e., severely impairing …

abstract arxiv attacks autonomous autonomous driving backdoor blindness classification cs.cv detection detectors driving focus however locations object overload paper safety safety-critical systems tasks type vulnerabilities