@article{1214cb72-3b4b-4ae8-bb0a-f7d7f1662417,
author = {Michał  Zając , Konrad Żołna , Negar  Rostamzadeh , Pedro O.  Pinheiro },
title = {Adversarial Framing for Image and Video Classification},
journal = {Schedae Informaticae},
volume = {2018},
number = {Volume 27},
year = {2018},
issn = {1732-3916},
pages = {155-164},keywords = {adversarial samples; convolutional neural networks; classification},
abstract = {Neural networks are prone to adversarial attacks. In general, such attacks deteriorate the quality of the input by either slightly modifying most of its pixels, or by occluding it with a patch. In this paper, we propose a method that keeps the image unchanged and only adds an adversarial framing on the border of the image. We show empirically that our method is able to successfully attack state-of-the-art methods on both image and video classification problems. Notably, the proposed method results in a universal attack which is very fast at test time.},
doi = {10.4467/20838476SI.18.012.10417},
url = {https://ejournals.eu/en/journal/schedae-informaticae/article/adversarial-framing-for-image-and-video-classification}
}