数字水印
计算机科学
人工智能
算法
模式识别(心理学)
数学
复制保护
稳健性(进化)
数据建模
数据挖掘
理论计算机科学
方案(数学)
噪音(视频)
图像分割
图像处理
图像(数学)
多维系统
信息隐藏
数据压缩
计算机视觉
作者
Peian Chen,Wenkang Su,Jian Zhang,Rui Duan,Tong Zhu,Zhili Zhou
标识
DOI:10.1109/tcsvt.2026.3724144
摘要
In-generation watermarking has emerged as a promising solution for provenance tracing and copyright protection of AI-generated images. Existing watermarking methods for Latent Diffusion Models (LDMs), however, still face a critical challenge in balancing generative fidelity and robustness against severe attacks. In this paper, we propose MsLW, a unified watermarking framework for LDMs that propagates watermark signals throughout the late-stage denoising trajectory, enabling the watermark to become an intrinsic component of the image generation process. Specifically, we introduce a lightweight Latent Watermarking Network (LWN) that learns the generation of watermark residual in the denoised latent space, together with a Multi-step Watermark Embedding (MsWE) strategy for progressively injecting the learned identical watermark residual into not only the denoised latent variable but also the intermediate latent ones. To mitigate distribution disturbance caused by iterative watermark propagation, MsWE further incorporates a latent-space distribution alignment mechanism, as well as wavelet low-pass filtering to suppress perceptual artifacts. Extensive experiments demonstrate that MsLW consistently outperforms existing SOTA watermarking methods under both destructive and regenerative attacks while maintaining superior perceptual quality and generative fidelity. The source code and additional visualization results can be found at https://github.com/chenpian/MsLW.
科研通智能强力驱动
Strongly Powered by AbleSci AI