Year: 2018
Type: preprint
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. Source code can be found at https://github.com/zajaczajac/adv_framing .
Source: arXiv (Cornell University)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: Narodowym Centrum Nauki
OpenAlex ID: W2904925408