A Large-scale Multiple-objective Method for Black-box Attack against Object Detection

TL;DR

This paper introduces GARSDC, a novel large-scale multi-objective optimization method that significantly enhances black-box adversarial attacks on object detectors by efficiently optimizing over millions of variables.

Contribution

It proposes a new multi-objective attack framework with an improved genetic algorithm and gradient-based initialization, achieving better attack performance and efficiency.

Findings

Reduces mean Average Precision (mAP) by 12.0 on average.

Decreases query count by about 1000 times.

Outperforms state-of-the-art attack methods in experiments.

Abstract

Recent studies have shown that detectors based on deep models are vulnerable to adversarial examples, even in the black-box scenario where the attacker cannot access the model information. Most existing attack methods aim to minimize the true positive rate, which often shows poor attack performance, as another sub-optimal bounding box may be detected around the attacked bounding box to be the new true positive one. To settle this challenge, we propose to minimize the true positive rate and maximize the false positive rate, which can encourage more false positive objects to block the generation of new true positive bounding boxes. It is modeled as a multi-objective optimization (MOP) problem, of which the generic algorithm can search the Pareto-optimal. However, our task has more than two million decision variables, leading to low searching efficiency. Thus, we extend the standard Genetic Algorithm with Random Subset selection and Divide-and-Conquer, called GARSDC, which significantly improves the efficiency. Moreover, to alleviate the sensitivity to population quality in generic algorithms, we generate a gradient-prior initial population, utilizing the transferability between different detectors with similar backbones. Compared with the state-of-art attack methods, GARSDC decreases by an average 12.0 in the mAP and queries by about 1000 times in extensive experiments. Our codes can be found at https://github.com/LiangSiyuan21/ GARSDC.