一拍即合：1步逆推实现扩散水印高效提取
OSI: One-step Inversion Excels in Extracting Diffusion Watermarks
- 将水印提取转为可学习的符号分类任务，跳过复杂噪声重建
- 比多步逆推快20倍，准确率更高，水印容量翻倍
- 适配多种扩散模型与加密方案，通用性强
水印是保护扩散生成图像来源与版权的重要机制。无需训练的方法（如高斯着色）通过在扩散模型初始噪声中嵌入水印，对生成图像质量影响极小。然而，传统水印提取需多步扩散逆推以精确恢复初始噪声，计算成本高且耗时。为此，我们提出一步逆推（OSI），一种更快更准的高斯着色类水印提取方法。OSI将水印提取重构为可学习的符号分类问题，无需精确回归初始噪声。我们从扩散主干网络初始化OSI模型，并在合成的噪声-图像对上以符号分类目标进行微调。该方法仅用一步即可高效完成水印提取。大量实验表明，OSI显著优于多步逆推：速度快20倍，提取准确率更高，水印容量翻倍。在多种调度器、扩散主干和加密方案下均表现一致提升，验证了框架的通用性。
原文摘要 · Abstract (English)
Watermarking is an important mechanism for provenance and copyright protection of diffusion-generated images. Training-free methods, exemplified by Gaussian Shading, embed watermarks into the initial noise of diffusion models with negligible impact on the quality of generated images. However, extracting this type of watermark typically requires multi-step diffusion inversion to obtain precise initial noise, which is computationally expensive and time-consuming. To address this issue, we propose One-step Inversion (OSI), a significantly faster and more accurate method for extracting Gaussian Shading style watermarks. OSI reformulates watermark extraction as a learnable sign classification problem, which eliminates the need for precise regression of the initial noise. Then, we initialize the OSI model from the diffusion backbone and finetune it on synthesized noise-image pairs with a sign classification objective. In this manner, the OSI model is able to accomplish the watermark extraction efficiently in only one step. Our OSI substantially outperforms the multi-step diffusion inversion method: it is 20x faster, achieves higher extraction accuracy, and doubles the watermark payload capacity. Extensive experiments across diverse schedulers, diffusion backbones, and cryptographic schemes consistently show improvements, demonstrating the generality of our OSI framework.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。