Rel-Zero：利用图像块对不变性实现抗 AI 编辑的鲁棒零水印
Rel-Zero: Harnessing Patch-Pair Invariance for Robust Zero-Watermarking Against AI Editing
提出以图像块对之间的相对距离作为编辑不变特征来构造零水印，为视觉研究者在面对扩散编辑的内容溯源与篡改检测时提供了一个无需改动原图、可复用的结构一致性表征思路。
原文依据 the relational distance between patch pairs remains relatively invariant
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期基于扩散模型的图像编辑技术对数字视觉内容的真实性构成了重大威胁。传统基于嵌入的水印方法为保持鲁棒性往往引入可感知的扰动，不可避免地损害视觉保真度。与此同时，现有的零水印方法通常依赖全局图像特征，难以抵御复杂的篡改操作。在本工作中，我们揭示了一个关键观察：尽管单张图像块在基于 AI 的编辑过程中会发生显著变化，图像块对之间的关系距离却保持相对不变。基于这一性质，我们提出了关系零水印（Relational Zero-Watermarking，Rel-Zero），这是一个无需修改原始图像、而是从这些对编辑保持不变性的图像块关系中导出唯一零水印的新框架。通过将水印建立在内在结构一致性而非绝对外观之上，Rel-Zero 为内容认证提供了一种非侵入式且具有韧性的机制。大量实验表明，与先前的零水印方法相比，Rel-Zero 在多种编辑模型与篡改操作下均取得了显著提升的鲁棒性。
英文原文摘要 1230 字符
Recent advancements in diffusion-based image editing pose a significant threat to the authenticity of digital visual content. Traditional embedding-based watermarking methods often introduce perceptible perturbations to maintain robustness, inevitably compromising visual fidelity. Meanwhile, existing zero-watermarking approaches, typically relying on global image features, struggle to withstand sophisticated manipulations. In this work, we uncover a key observation: while individual image patches undergo substantial alterations during AI-based editing, the relational distance between patch pairs remains relatively invariant. Leveraging this property, we propose Relational Zero-Watermarking (Rel-Zero), a novel framework that requires no modification to the original image but derives a unique zero-watermark from these editing-invariant patch relations. By grounding the watermark in intrinsic structural consistency rather than absolute appearance, Rel-Zero provides a non-invasive yet resilient mechanism for content authentication. Extensive experiments demonstrate that Rel-Zero achieves substantially improved robustness across diverse editing models and manipulations compared to prior zero-watermarking approaches.