arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.08337cs.CVeess.IV
收藏

WMAdapter:为潜在扩散模型添加水印控制

WMAdapter: Adding WaterMark Control to Latent Diffusion Models

National University of Singapore(新加坡国立大学);Show Lab

Hai Ci, Yiren Song, Pei Yang, Jinheng Xie, Mike Zheng Shou

AI总结:

本文提出WMAdapter水印插件,通过轻量级上下文适配器与混合微调,在潜在扩散生成过程中嵌入用户指定水印,兼顾图像质量、灵活性与稳健性。

摘要:

水印对于保护AI生成图像的版权至关重要。我们提出WMAdapter,这是一种扩散模型水印插件,它接收用户指定的水印信息,并允许在扩散生成过程中无缝压印水印。WMAdapter高效且稳健,并高度重视高生成质量。为实现这一点,我们做出两项关键设计:(1)我们开发了一种轻量级的上下文适配器结构,能够从经过大规模预训练的事后水印模型中有效迁移知识。(2)我们引入额外的微调步骤,并设计混合微调策略,以进一步提升图像质量并消除微小伪影。实证结果表明,WMAdapter具备强大的灵活性、出色的图像生成质量以及具有竞争力的水印稳健性。

英文摘要:

Watermarking is crucial for protecting the copyright of AI-generated images. We propose WMAdapter, a diffusion model watermark plugin that takes user-specified watermark information and allows for seamless watermark imprinting during the diffusion generation process. WMAdapter is efficient and robust, with a strong emphasis on high generation quality. To achieve this, we make two key designs: (1) We develop a contextual adapter structure that is lightweight and enables effective knowledge transfer from heavily pretrained post-hoc watermarking models. (2) We introduce an extra finetuning step and design a hybrid finetuning strategy to further improve image quality and eliminate tiny artifacts. Empirical results demonstrate that WMAdapter offers strong flexibility, exceptional image generation quality and competitive watermark robustness.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑