Naturalistic Physical Adversarial Patch for Object Detectors

Most prior works on physical adversarial attacks mainly focus on the attack performance but seldom enforce any restrictions over the appearance of the generated adversarial patches. This leads to conspicuous and attention-grabbing patterns for the generated patches which can be easily identified by...

Full description

Saved in:
Bibliographic Details
Main Authors: Hu, Yu-Chih-Tuan, Chen, Jun-Cheng, Kung, Bo-Han, Hua, Kai-Lung, Tan, Daniel Stanley
Format: Conference Proceeding
Language:English
Subjects:
Adversarial learning
Computational modeling
Computer vision
Detection and localization in 2D and 3D
Detectors
Generative adversarial networks
Image synthesis
Manifolds
Quality assessment
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Most prior works on physical adversarial attacks mainly focus on the attack performance but seldom enforce any restrictions over the appearance of the generated adversarial patches. This leads to conspicuous and attention-grabbing patterns for the generated patches which can be easily identified by humans. To address this issue, we pro-pose a method to craft physical adversarial patches for object detectors by leveraging the learned image manifold of a pretrained generative adversarial network (GAN) (e.g., BigGAN and StyleGAN) upon real-world images. Through sampling the optimal image from the GAN, our method can generate natural looking adversarial patches while maintaining high attack performance. With extensive experiments on both digital and physical domains and several independent subjective surveys, the results show that our proposed method produces significantly more realistic and natural looking patches than several state-of-the-art base-lines while achieving competitive attack performance. 1
ISSN:2380-7504
DOI:10.1109/ICCV48922.2021.00775