Relevance Attack on Detectors

Document Type

Article

Publication Title

Pattern Recognition

Abstract

This paper focuses on high-transferable adversarial attacks on detectors, which are hard to attack in a black-box manner, because of their multiple-output characteristics and the diversity across architectures. To pursue a high attack transferability, one plausible way is to find a common property across detectors, which facilitates the discovery of common weaknesses. We are the first to suggest that the relevance map from interpreters for detectors is such a property. Based on it, we design a Relevance Attack on Detectors (RAD), which achieves a state-of-the-art transferability, exceeding existing results by above 20%. On MS COCO, the detection mAPs for all 8 black-box architectures are more than halved and the segmentation mAPs are also significantly influenced. Given the great transferability of RAD, we generate the first adversarial dataset for object detection and instance segmentation, i.e., Adversarial Objects in COntext (AOCO), which helps to quickly evaluate and improve the robustness of detectors.

DOI

10.1016/j.patcog.2021.108491

Publication Date

4-1-2022

Keywords

Adversarial attack, Attack transferability, Black-box attack, Interpreters, Object detection, Relevance map

Comments

IR deposit conditions:

  • OA (accepted version) - pathway b
  • 24 months embargo
  • Must link to publisher version with DOI

Link to Full Text

Share

COinS