| Référence : Evasion Attack STeganography: Turning Vulnerability Of Machine Learning ToAdversarial... |
| Périodiques scientifiques : Article | |||
| Ingénierie, informatique & technologie : Sciences informatiques | |||
| Security, Reliability and Trust | |||
| http://hdl.handle.net/10993/47832 | |||
| Evasion Attack STeganography: Turning Vulnerability Of Machine Learning ToAdversarial Attacks Into A Real-world Application | |
| anglais | |
Ghamizi, Salah [University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS) >] | |
Cordy, Maxime [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SerVal >] | |
Papadakis, Mike [University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS) >] | |
Le Traon, Yves [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SerVal >] | |
| 11-oct-2021 | |
| Proceedings of International Conference on Computer Vision 2021 | |
| Oui | |
| Internationale | |
| [en] Adversarial Attacks ; Steganography ; Watermarking | |
| [en] Evasion Attacks have been commonly seen as a weakness of Deep Neural Networks. In this paper, we flip the paradigm and envision this vulnerability as a useful application.
We propose EAST, a new steganography and watermarking technique based on multi-label targeted evasion attacks. Our results confirm that our embedding is elusive; it not only passes unnoticed by humans, steganalysis methods, and machine-learning detectors. In addition, our embedding is resilient to soft and aggressive image tampering (87% recovery rate under jpeg compression). EAST outperforms existing deep-learning-based steganography approaches with images that are 70% denser and 73% more robust and supports multiple datasets and architectures. | |
| Chercheurs | |
| http://hdl.handle.net/10993/47832 |
| Document(s) associé(s) à cette référence : | ||||||||||||||
|
Document(s) en texte intégral :
| ||||||||||||||
Tous les documents dans ORBilu sont protégés par une licence d'utilisation.