Toward Robust Neural Image Compression: Adversarial Attack and Model Finetuning
IEEE Transactions on Circuits and Systems for Video Technology · 2023-12 · 33 (12) · 7842-7856
DOI: 10.1109/tcsvt.2023.3276442 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2023-12. Verfasst von Tong Chen und 1 weiteren Autor:innen. Die Arbeit wurde bisher 47-mal zitiert (Crossref) und nennt 68 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Deep neural network-based image compression has been extensively studied. However, the model robustness which is crucial to practical application is largely overlooked. We propose to examine the robustness of prevailing learned image compression models by injecting negligible adversarial perturbation into the original source image. Severe distortion in decoded reconstruction reveals the general vulnerability in existing methods regardless of their settings (e.g., network architecture, loss function, quality scale). A variety of defense strategies including geometric self-ensemble based pre-processing, and adversarial training, are investigated against the adversarial attack to improve the model’s robustness. Later the defense efficiency is further exemplified in real-life image recompression case studies. Overall, our methodology is simple, effective, and generalizable, making it attractive for developing robust learned image compression solutions. All materials are made publicly accessible at https://njuvision.github.io/RobustNIC for reproducible research.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Chen, T., & Ma, Z. (2023). Toward Robust Neural Image Compression: Adversarial Attack and Model Finetuning. IEEE Transactions on Circuits and Systems for Video Technology, 33(12), 7842-7856. https://doi.org/10.1109/tcsvt.2023.3276442
Vancouver
Chen T, Ma Z. Toward Robust Neural Image Compression: Adversarial Attack and Model Finetuning. IEEE Transactions on Circuits and Systems for Video Technology. 2023;33(12):7842-7856. doi:10.1109/tcsvt.2023.3276442
BibTeX
@article{Chen2023,
author = {Chen, Tong and Ma, Zhan},
title = {Toward Robust Neural Image Compression: Adversarial Attack and Model Finetuning},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2023},
volume = {33},
number = {12},
pages = {7842--7856},
doi = {10.1109/tcsvt.2023.3276442}
}
EUVIMED