On the Real-World Adversarial Robustness of Real-Time Semantic Segmentation Models for Autonomous Driving

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rossolini, Giulio, Nesti, Federico, D'Amico, Gianluca, Nair, Saasha, Biondi, Alessandro, Buttazzo, Giorgio
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912577073709056
author Rossolini, Giulio
Nesti, Federico
D'Amico, Gianluca
Nair, Saasha
Biondi, Alessandro
Buttazzo, Giorgio
author_facet Rossolini, Giulio
Nesti, Federico
D'Amico, Gianluca
Nair, Saasha
Biondi, Alessandro
Buttazzo, Giorgio
contents The existence of real-world adversarial examples (commonly in the form of patches) poses a serious threat for the use of deep learning models in safety-critical computer vision tasks such as visual perception in autonomous driving. This paper presents an extensive evaluation of the robustness of semantic segmentation models when attacked with different types of adversarial patches, including digital, simulated, and physical ones. A novel loss function is proposed to improve the capabilities of attackers in inducing a misclassification of pixels. Also, a novel attack strategy is presented to improve the Expectation Over Transformation method for placing a patch in the scene. Finally, a state-of-the-art method for detecting adversarial patch is first extended to cope with semantic segmentation models, then improved to obtain real-time performance, and eventually evaluated in real-world scenarios. Experimental results reveal that, even though the adversarial effect is visible with both digital and real-world attacks, its impact is often spatially confined to areas of the image around the patch. This opens to further questions about the spatial robustness of real-time semantic segmentation models.
format Preprint
id arxiv_https___arxiv_org_abs_2201_01850
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle On the Real-World Adversarial Robustness of Real-Time Semantic Segmentation Models for Autonomous Driving
Rossolini, Giulio
Nesti, Federico
D'Amico, Gianluca
Nair, Saasha
Biondi, Alessandro
Buttazzo, Giorgio
Computer Vision and Pattern Recognition
Artificial Intelligence
The existence of real-world adversarial examples (commonly in the form of patches) poses a serious threat for the use of deep learning models in safety-critical computer vision tasks such as visual perception in autonomous driving. This paper presents an extensive evaluation of the robustness of semantic segmentation models when attacked with different types of adversarial patches, including digital, simulated, and physical ones. A novel loss function is proposed to improve the capabilities of attackers in inducing a misclassification of pixels. Also, a novel attack strategy is presented to improve the Expectation Over Transformation method for placing a patch in the scene. Finally, a state-of-the-art method for detecting adversarial patch is first extended to cope with semantic segmentation models, then improved to obtain real-time performance, and eventually evaluated in real-world scenarios. Experimental results reveal that, even though the adversarial effect is visible with both digital and real-world attacks, its impact is often spatially confined to areas of the image around the patch. This opens to further questions about the spatial robustness of real-time semantic segmentation models.
title On the Real-World Adversarial Robustness of Real-Time Semantic Segmentation Models for Autonomous Driving
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2201.01850