Metraipolis › archive › dc_9df4f71e6dd4

Where Does the Watermark Hide? Push-Pull Disentanglement for Invisible Watermark Removal — arXiv:2609.31722

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31722
License (as archived)
CC0-1.0
Archived
2026-10-07T23:50:32.281Z
SHA-256 of the archived text
c7360730311a7b8dcdc759308cc651e1d8fb8847ecc722ec15e92a7a10bc0201
Stable URL of this page
https://metraipolis.com/documento/dc_9df4f71e6dd4

How to cite

Where Does the Watermark Hide? Push-Pull Disentanglement for Invisible Watermark Removal — arXiv:2609.31722 — arXiv (en), https://arxiv.org/abs/2609.31722 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9df4f71e6dd4 (Metraipolis archive dc_9df4f71e6dd4, sha256 c7360730311a7b8dcdc759308cc651e1d8fb8847ecc722ec15e92a7a10bc0201)
BibTeX
@misc{dc9df4f71e6dd4,
  title = {Where Does the Watermark Hide? Push-Pull Disentanglement for Invisible Watermark Removal — arXiv:2609.31722},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31722},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9df4f71e6dd4; sha256 c7360730311a7b8dcdc759308cc651e1d8fb8847ecc722ec15e92a7a10bc0201},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31722. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fixed image distortions do not cover an attacker that learns from paired clean and watermarked images. We study this paired-training threat with single-image inference: deployment uses neither the clean reference nor the watermark key, payload, or decoder. An encoder maps each image to a structural latent $g$ and an auxiliary residual latent $u$. Push supervision reconstructs the watermarked image from $D(g_w,u_w)$. Pull supervision trains the zero-auxiliary output $D(A_g(g_w;k),0)$ toward the paired clean image. At $k=1.10,u=0$, the four-method sweep gives an average BER of $0.3958$, PSNR of $31.07$ dB, and SSIM of $0.9554$. Restoring $u$ from $0$ to $0.15$ moves average BER from $0.3893$ to $0.3357$, while PSNR falls from $30.99$ to $28.23$ dB. The intervention supports decoder dependence on the auxiliary input in the evaluated setting. The accompanying theory is a conditional, post-hoc account of this behavior rather than an experimentally verified information-relocation result.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.31722 — 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 2026-10-07T23:29:26.899Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).