Universal Adversarial Perturbations Against Semantic Image Segmentation

Bibliographic Details
Authors and Corporations: Metzen, Jan Hendrik, Kumar, Mummadi Chaithanya, Brox, Thomas, Fischer, Volker
Title: Universal Adversarial Perturbations Against Semantic Image Segmentation
In: 2017 IEEE International Conference on Computer Vision (ICCV), 2017, p. 2774-2783
Physical Description:2774-2783
ISSN/ISBN: 978-1-5386-1032-9
Summary:While deep learning is remarkably successful on perceptual tasks, it was also shown to be vulnerable to adversarial perturbations of the input. These perturbations denote noise added to the input that was generated specifically to fool the system while being quasi-imperceptible for humans. More severely, there even exist universal perturbations that are input-agnostic but fool the network on the majority of inputs. While recent work has focused on image classification, this work proposes attacks against semantic image segmentation: we present an approach for generating (universal) adversarial perturbations that make the network yield a desired target segmentation as output. We show empirically that there exist barely perceptible universal noise patterns which result in nearly the same predicted segmentation for arbitrary inputs. Furthermore, we also show the existence of universal noise which removes a target class (e.g., all pedestrians) from the segmentation while leaving the segmentation mostly unchanged otherwise.
Type of Resource:E-Article
Source:IEEE Xplore Library
Language: English