Apr 19, 2017 � This work proposes attacks against semantic image segmentation: we present an approach for generating (universal) adversarial perturbations.

In this paper, we investigate the ef- fect of adversarial attacks on tasks involving a localization component, more specifically: semantic image segmenta- tion.

While deep learning is remarkably successful on percep- tual tasks, it was also shown to be vulnerable to adversar- ial perturbations of the input.

This work presents an approach for generating (universal) adversarial perturbations that make the network yield a desired target segmentation as output.

This work proposes attacks against semantic image segmentation: we present an approach for generating (universal) adversarial perturbations.

The first category consists of common attacks that were directly transferred from image classification to semantic segmentation models, making use of the�...

Apr 19, 2017 � While recent work has focused on image classification, this work proposes attacks against semantic image segmentation: we present an approach�...

This paper presents what to their knowledge is the first rigorous evaluation of adversarial attacks on modern semantic segmentation models.

Oct 26, 2016 � We propose a systematic algorithm for computing universal perturbations, and show that state-of-the-art deep neural networks are highly vulnerable to such�...

We propose a Universal Dense Object Suppression (U-DOS) algorithm to derive the universal adversarial perturbations against object detection.