AVA: Adversarial Vignetting Attack against visual recognition
Vignetting is an inherent imaging phenomenon within almost all optical systems, showing as a radial intensity darkening toward the corners of an image. Since it is a common effect for photography and usually appears as a slight intensity variation, people usually regard it as a part of a photo and w...
Saved in:
Main Authors: | , , , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2021
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/7087 https://ink.library.smu.edu.sg/context/sis_research/article/8090/viewcontent/0145.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-8090 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-80902022-04-07T07:39:00Z AVA: Adversarial Vignetting Attack against visual recognition TIAN, Binyu JUEFEI-XU, Felix GUO, Qing XIE, Xiaofei LI, Xiaohong LIU, Yang Vignetting is an inherent imaging phenomenon within almost all optical systems, showing as a radial intensity darkening toward the corners of an image. Since it is a common effect for photography and usually appears as a slight intensity variation, people usually regard it as a part of a photo and would not even want to post-process it. Due to this natural advantage, in this work, we study the vignetting from a new viewpoint, i.e., adversarial vignetting attack (AVA), which aims to embed intentionally misleading information into the vignetting and produce a natural adversarial example without noise patterns. This example can fool the state-of-the-art deep convolutional neural networks (CNNs) but is imperceptible to human. To this end, we first propose the radial-isotropic adversarial vignetting attack (RI-AVA) based on the physical model of vignetting, where the physical parameters (e.g., illumination factor and focal length) are tuned through the guidance of target CNN models. To achieve higher transferability across different CNNs, we further propose radial-anisotropic adversarial vignetting attack (RA-AVA) by allowing the effective regions of vignetting to be radial-anisotropic and shape-free. Moreover, we propose the geometry-aware level-set optimization method to solve the adversarial vignetting regions and physical parameters jointly. We validate the proposed methods on three popular datasets, i.e., DEV, CIFAR10, and Tiny ImageNet, by attacking four CNNs, e.g., ResNet50, EfficientNet-B0, DenseNet121, and MobileNet-V2, demonstrating the advantages of our methods over baseline methods on both transferability and image quality. 2021-08-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7087 info:doi/10.24963/ijcai.2021/145 https://ink.library.smu.edu.sg/context/sis_research/article/8090/viewcontent/0145.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Computer Vision: Recognition: Detection Categorization Indexing Matching Retrieval Semantic Interpretation OS and Networks Software Engineering |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Computer Vision: Recognition: Detection Categorization Indexing Matching Retrieval Semantic Interpretation OS and Networks Software Engineering |
spellingShingle |
Computer Vision: Recognition: Detection Categorization Indexing Matching Retrieval Semantic Interpretation OS and Networks Software Engineering TIAN, Binyu JUEFEI-XU, Felix GUO, Qing XIE, Xiaofei LI, Xiaohong LIU, Yang AVA: Adversarial Vignetting Attack against visual recognition |
description |
Vignetting is an inherent imaging phenomenon within almost all optical systems, showing as a radial intensity darkening toward the corners of an image. Since it is a common effect for photography and usually appears as a slight intensity variation, people usually regard it as a part of a photo and would not even want to post-process it. Due to this natural advantage, in this work, we study the vignetting from a new viewpoint, i.e., adversarial vignetting attack (AVA), which aims to embed intentionally misleading information into the vignetting and produce a natural adversarial example without noise patterns. This example can fool the state-of-the-art deep convolutional neural networks (CNNs) but is imperceptible to human. To this end, we first propose the radial-isotropic adversarial vignetting attack (RI-AVA) based on the physical model of vignetting, where the physical parameters (e.g., illumination factor and focal length) are tuned through the guidance of target CNN models. To achieve higher transferability across different CNNs, we further propose radial-anisotropic adversarial vignetting attack (RA-AVA) by allowing the effective regions of vignetting to be radial-anisotropic and shape-free. Moreover, we propose the geometry-aware level-set optimization method to solve the adversarial vignetting regions and physical parameters jointly. We validate the proposed methods on three popular datasets, i.e., DEV, CIFAR10, and Tiny ImageNet, by attacking four CNNs, e.g., ResNet50, EfficientNet-B0, DenseNet121, and MobileNet-V2, demonstrating the advantages of our methods over baseline methods on both transferability and image quality. |
format |
text |
author |
TIAN, Binyu JUEFEI-XU, Felix GUO, Qing XIE, Xiaofei LI, Xiaohong LIU, Yang |
author_facet |
TIAN, Binyu JUEFEI-XU, Felix GUO, Qing XIE, Xiaofei LI, Xiaohong LIU, Yang |
author_sort |
TIAN, Binyu |
title |
AVA: Adversarial Vignetting Attack against visual recognition |
title_short |
AVA: Adversarial Vignetting Attack against visual recognition |
title_full |
AVA: Adversarial Vignetting Attack against visual recognition |
title_fullStr |
AVA: Adversarial Vignetting Attack against visual recognition |
title_full_unstemmed |
AVA: Adversarial Vignetting Attack against visual recognition |
title_sort |
ava: adversarial vignetting attack against visual recognition |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2021 |
url |
https://ink.library.smu.edu.sg/sis_research/7087 https://ink.library.smu.edu.sg/context/sis_research/article/8090/viewcontent/0145.pdf |
_version_ |
1770576209697046528 |