FAQ
logo of Jagiellonian University in Krakow

Adversarial Framing for Image and Video Classification

Publication date: 2018

Schedae Informaticae, 2018, Volume 27, pp. 155 - 164

https://doi.org/10.4467/20838476SI.18.012.10417

Authors

,
Michał Zając
Jagiellonian University in Kraków
https://orcid.org/0000-0001-9096-8258 Orcid
All publications →
,
Konrad Żołna
Element AI, Montreal, Canada
Jagiellonian University in Kraków
All publications →
,
Negar Rostamzadeh
Element AI, Montreal, Canada
All publications →
Pedro O. Pinheiro
Element AI, Montreal, Canada
All publications →

Titles

Adversarial Framing for Image and Video 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.

References


Information

Information: Schedae Informaticae, 2018, Volume 27, pp. 155 - 164

Article type: Original article

Titles:

Polish:

Adversarial Framing for Image and Video Classification

English:

Adversarial Framing for Image and Video Classification

Authors

https://orcid.org/0000-0001-9096-8258

Michał Zając
Jagiellonian University in Kraków
https://orcid.org/0000-0001-9096-8258 Orcid
All publications →

Jagiellonian University in Kraków

Element AI, Montreal, Canada

Jagiellonian University in Kraków

Element AI, Montreal, Canada

Element AI, Montreal, Canada

Published at: 2018

Article status: Open

Licence: CC BY-NC-ND  licence icon

Percentage share of authors:

Michał Zając (Author) - 25%
Konrad Żołna (Author) - 25%
Negar Rostamzadeh (Author) - 25%
Pedro O. Pinheiro (Author) - 25%

Article corrections:

-

Publication languages:

English

View count: 1273

Number of downloads: 0

<p> Adversarial Framing for Image and Video Classification</p>