Metraipolis › archive › dc_9aeabc9ca893
What Limits Robustness in Deep Image Watermarking: An Analysis of Mechanisms and Their Scaling Across Capacities — arXiv:2609.01050
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.01050
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T06:46:43.942Z
- SHA-256 of the archived text
74ebacf55d9fee4c93c105954a251887e6ecd6cbeb89d07817cf1bf7e9b0418f- Stable URL of this page
https://metraipolis.com/documento/dc_9aeabc9ca893
How to cite
What Limits Robustness in Deep Image Watermarking: An Analysis of Mechanisms and Their Scaling Across Capacities — arXiv:2609.01050 — arXiv (en), https://arxiv.org/abs/2609.01050 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9aeabc9ca893 (Metraipolis archive dc_9aeabc9ca893, sha256 74ebacf55d9fee4c93c105954a251887e6ecd6cbeb89d07817cf1bf7e9b0418f)
BibTeX
@misc{dc9aeabc9ca893,
title = {What Limits Robustness in Deep Image Watermarking: An Analysis of Mechanisms and Their Scaling Across Capacities — arXiv:2609.01050},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.01050},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9aeabc9ca893; sha256 74ebacf55d9fee4c93c105954a251887e6ecd6cbeb89d07817cf1bf7e9b0418f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01050. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Robustness remains the principal open problem in deep image watermarking, and what limits it becomes sharper as payload grows. This paper asks whether capacity is itself the limit or only makes other limits visible, and answers in two parts. The first organizes the distortions a watermark must survive and the strategies developed to resist them, ordering each by the axis that governs it: payload capacity for the distortions, differentiability for the strategies. The second identifies and measures three mechanisms that limit robustness in schemes mapping the payload onto a spatial block grid with extraction trained separately from a frozen embedder: desynchronization of the payload grid, the resistance of codec-induced distortion to training, and the narrowing of the usable embedding-strength window. Payloads from 64 to 16384 bits are measured, well beyond the range those strategies address. Training the extraction stage against a codec proves not merely ineffective but harmful, degrading the reading at the operating points used in training. The limits follow the class of distortion rather than capacity itself, and none is removed by further training on the extraction side, because all three arise before extraction. An evaluation protocol making claims of generalization verifiable is also contributed. The conclusions are properties of a class of designs rather than of one implementation. Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CR). Categorie arXiv: Cryptography and Security (cs.CR). Pagina del lavoro: https://arxiv.org/abs/2609.01050 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T06:38:14.370Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).