Universal Physical Camouflage Attacks on Object Detectors
Preprint
Abstract
In this paper, we study physical adversarial attacks on object detectors in the wild. Previous works mostly craft instance-dependent perturbations only for rigid or planar objects. To this end, we propose to learn an adversarial pattern to effectively attack all instances belonging to the same object category, referred to as Universal Physical Camouflage Attack (UPC). Concretely, UPC crafts camouflage by jointly fooling the region proposal network, as well as misleading the classifier and the regressor to output errors. In order to make UPC effective for non-rigid or non-planar objects, we introduce a set of transformations for mimicking deformable properties. We additionally impose optimization constraint to make generated patterns look natural to human observers. To fairly evaluate the effectiveness of different physical-world attacks, we present the first standardized virtual database, AttackScenes, which simulates the real 3D world in a controllable and reproducible environment. Extensive experiments suggest the superiority of our proposed UPC compared with existing physical adversarial attackers not only in virtual environments (AttackScenes), but also in real-world physical environments. Code and dataset are available at https://mesunhlf.github.io/index_physical.html.Keywords
All Related Versions
This publication has 4 references indexed in Scilit:
- Humans can decipher adversarial imagesNature Communications, 2019
- TextBugger: Generating Adversarial Text Against Real-world Applications2019
- Adversarial Examples in the Physical World2018
- Scale-aware Fast R-CNN for Pedestrian DetectionIEEE Transactions on Multimedia, 2017