RAIN：面向语义水印提取的区域感知反演网络

View PDF HTML (experimental)

Abstract:Semantic watermarks for diffusion models embed ownership information into the generative process while preserving perceptual quality, but Gaussian-Shading extraction conventionally requires multi-step diffusion inversion to recover the initial noise. Recent one-step methods show that this cost can be reduced substantially. We study this problem through extended flow matching and conditional regression. The key observation is that, near the high-SNR image endpoint, recovering a useful noise statistic given by the first-step output of the extended flow matching in the high-SNR regime is much simpler than reconstructing the full inverse trajectory, and Gaussian Shading only requires the recovered latent to remain in the correct watermark decision region. Based on this observation, we propose a lightweight, prompt-free extractor that decomposes endpoint recovery into an image-like anchor and a noise-oriented residual, which increases the capability of the model to utilize GPU parallel computation. The resulting method avoids iterative inversion and repeated evaluation of a diffusion-scale U-Net, providing an efficient one-step extraction pipeline with a concise theoretical interpretation. The computational cost of extracting noise is lower than that of both OSI and FARI. The github repo is there: this https URL

Comments:	Accepted to the BMVC 2026 Workshop (TrustVis) as an Oral Presentation
Subjects:	Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2609.14856 [cs.CV]
	(or arXiv:2609.14856v3 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2609.14856 arXiv-issued DOI via DataCite

Submission history

From: Zilai Li [view email]
[v1] Mon, 14 Sep 2026 00:07:17 UTC (273 KB)
[v2] Tue, 22 Sep 2026 23:58:40 UTC (274 KB)
[v3] Thu, 12 Nov 2026 06:44:40 UTC (275 KB)