Adversarial Images through Stega Glasses - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year : 2020

Adversarial Images through Stega Glasses

Benoît Bonnet
  • Function : Author
  • PersonId : 1077796
Teddy Furon

Abstract

This paper explores the connection between steganography and adversarial images. On the one hand, ste-ganalysis helps in detecting adversarial perturbations. On the other hand, steganography helps in forging adversarial perturbations that are not only invisible to the human eye but also statistically undetectable. This work explains how to use these information hiding tools for attacking or defending computer vision image classification. We play this cat and mouse game with state-of-art classifiers, steganalyzers, and steganographic embedding schemes. It turns out that steganography helps more the attacker than the defender.
Fichier principal
Vignette du fichier
paper.pdf (707.79 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02946732 , version 1 (26-09-2020)

Identifiers

Cite

Benoît Bonnet, Teddy Furon, Patrick Bas. Adversarial Images through Stega Glasses. 2020. ⟨hal-02946732⟩
28 View
39 Download

Altmetric

Share

Gmail Facebook X LinkedIn More