On Evaluating the Adversarial Robustness of Semantic Segmentation Models
Achieving robustness against adversarial input perturbation is an important and intriguing problem in machine learning. In the area of semantic image segmentation, a number of adversarial training approaches have been proposed as a defense against adversarial perturbation, but the methodology of evaluating the robustness of the models is still lacking, compared to image classification. Here, we demonstrate that, just like in image classification, it is important to evaluate the models over several different and hard attacks. We propose a set of gradient based iterative attacks and show that it is essential to perform a large number of iterations. We include attacks against the internal representations of the models as well. We apply two types of attacks: maximizing the error with a bounded perturbation, and minimizing the perturbation for a given level of error. Using this set of attacks, we show for the first time that a number of models in previous work that are claimed to be robust are in fact not robust at all. We then evaluate simple adversarial training algorithms that produce reasonably robust models even under our set of strong attacks. Our results indicate that a key design decision to achieve any robustness is to use only adversarial examples during training. However, this introduces a trade-off between robustness and accuracy.
Code (0)
등록된 구현이 없습니다.
Tasks
Adversarial Robustnessimage-classificationImage ClassificationImage SegmentationSemantic SegmentationSimilar Papers 제목 키워드 기반
Evaluating the Robustness of Off-Road Autonomous Driving Segmentation against Adversarial Attacks: A Dataset-Centric analysis
This study investigates the vulnerability of semantic segmentation models to adversarial input perturbations, in the domain of off- road autonomous driving. Despite good performance in generic conditions, the state-of-th…
Autonomous DrivingSemantic SegmentationSegPGD: An Effective and Efficient Adversarial Attack for Evaluating and Boosting Segmentation Robustness
Deep neural network-based image classifications are vulnerable to adversarial perturbations. The image classifications can be easily fooled by adding artificial small and imperceptible perturbations to input images. As o…
Adversarial AttackSegmentationSemantic SegmentationSPARTA: Evaluating Reasoning Segmentation Robustness through Black-Box Adversarial Paraphrasing in Text Autoencoder Latent Space
Multimodal large language models (MLLMs) have shown impressive capabilities in vision-language tasks such as reasoning segmentation, where models generate segmentation masks based on textual queries. While prior work has…
Reinforcement LearningOn Evaluating Adversarial Robustness of Volumetric Medical Segmentation Models
Volumetric medical segmentation models have achieved significant success on organ and tumor-based segmentation tasks in recent years. However, their vulnerability to adversarial attacks remains largely unexplored, raisin…
Adversarial RobustnessMambaSegmentationSingle-Step Adversarial Training for Semantic Segmentation
Even though deep neural networks succeed on many different tasks including semantic segmentation, they lack on robustness against adversarial examples. To counteract this exploit, often adversarial training is used. Howe…
SegmentationSemantic Segmentation