On the Robustness of Semantic Segmentation Models to Adversarial Attacks
This work addresses the vulnerability of semantic segmentation models to adversarial attacks, which is crucial for safety-critical applications like autonomous driving, though it is incremental as it extends existing adversarial attack research to a more complex task.
The paper tackles the problem of adversarial attacks on semantic segmentation models, conducting the first rigorous evaluation on large-scale datasets and showing that observations from classification tasks do not always apply, with findings on model robustness aiding safety-critical applications.
Deep Neural Networks (DNNs) have demonstrated exceptional performance on most recognition tasks such as image classification and segmentation. However, they have also been shown to be vulnerable to adversarial examples. This phenomenon has recently attracted a lot of attention but it has not been extensively studied on multiple, large-scale datasets and structured prediction tasks such as semantic segmentation which often require more specialised networks with additional components such as CRFs, dilated convolutions, skip-connections and multiscale processing. In this paper, we present what to our knowledge is the first rigorous evaluation of adversarial attacks on modern semantic segmentation models, using two large-scale datasets. We analyse the effect of different network architectures, model capacity and multiscale processing, and show that many observations made on the task of classification do not always transfer to this more complex task. Furthermore, we show how mean-field inference in deep structured models, multiscale processing (and more generally, input transformations) naturally implement recently proposed adversarial defenses. Our observations will aid future efforts in understanding and defending against adversarial examples. Moreover, in the shorter term, we show how to effectively benchmark robustness and show which segmentation models should currently be preferred in safety-critical applications due to their inherent robustness.