Saved in:
Bibliographic Details
Main Authors: Bayer, Jens, Becker, Stefan, Münch, David, Arens, Michael
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.10963
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913616778756096
author Bayer, Jens
Becker, Stefan
Münch, David
Arens, Michael
author_facet Bayer, Jens
Becker, Stefan
Münch, David
Arens, Michael
contents Adversarial patches are still a simple yet powerful white box attack that can be used to fool object detectors by suppressing possible detections. The patches of these so-called evasion attacks are computational expensive to produce and require full access to the attacked detector. This paper addresses the problem of computational expensiveness by analyzing 375 generated patches, calculating the principal components of these and show, that linear combinations of the resulting "eigenpatches" can be used to fool object detections successfully.
format Preprint
id arxiv_https___arxiv_org_abs_2306_10963
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Eigenpatches -- Adversarial Patches from Principal Components
Bayer, Jens
Becker, Stefan
Münch, David
Arens, Michael
Computer Vision and Pattern Recognition
Adversarial patches are still a simple yet powerful white box attack that can be used to fool object detectors by suppressing possible detections. The patches of these so-called evasion attacks are computational expensive to produce and require full access to the attacked detector. This paper addresses the problem of computational expensiveness by analyzing 375 generated patches, calculating the principal components of these and show, that linear combinations of the resulting "eigenpatches" can be used to fool object detections successfully.
title Eigenpatches -- Adversarial Patches from Principal Components
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2306.10963