The Translucent Patch: A Physical and Universal Attack on Object Detectors

TL;DR

This paper introduces a contactless, translucent physical patch placed on a camera lens that effectively fools object detectors by hiding specific target objects without affecting the detection of other classes, advancing physical adversarial attack methods.

Contribution

The paper presents a novel contactless, translucent patch for physical attacks on object detectors, effective in hiding target objects while preserving detection of other classes.

Findings

Hides 42.27% of stop sign instances in experiments.

Maintains nearly 80% detection rate for other classes.

Works on state-of-the-art autonomous driving models.

Abstract

Physical adversarial attacks against object detectors have seen increasing success in recent years. However, these attacks require direct access to the object of interest in order to apply a physical patch. Furthermore, to hide multiple objects, an adversarial patch must be applied to each object. In this paper, we propose a contactless translucent physical patch containing a carefully constructed pattern, which is placed on the camera's lens, to fool state-of-the-art object detectors. The primary goal of our patch is to hide all instances of a selected target class. In addition, the optimization method used to construct the patch aims to ensure that the detection of other (untargeted) classes remains unharmed. Therefore, in our experiments, which are conducted on state-of-the-art object detection models used in autonomous driving, we study the effect of the patch on the detection of both the selected target class and the other classes. We show that our patch was able to prevent the detection of 42.27% of all stop sign instances while maintaining high (nearly 80%) detection of the other classes.