CVCRLGAug 16, 2020

Relevance Attack on Detectors

arXiv:2008.06822v418 citations
Originality Highly original
AI Analysis

This addresses the robustness problem for object detection systems, with incremental novelty in using relevance maps for attacks.

The paper tackles the challenge of creating high-transferability adversarial attacks on object detectors by exploiting relevance maps from interpreters, achieving state-of-the-art transferability with over 20% improvement and halving detection mAPs on MS COCO for black-box architectures.

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.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes