Conceptwm: A Diffusion Model Watermark for Concept Protection

The personalization techniques of diffusion models succeed in generating specific concepts but also pose threats to copyright protection and illegal use. Model Watermarking is an effective method to prevent the unauthorized use of subject-driven or style-driven image generation, safeguarding concept copyrights. However, under the goal of concept-oriented protection, current watermarking schemes typically add watermarks to all images rather than applying them in a refined manner targeted at specific concepts. Additionally, the personalization techniques of diffusion models can easily remove watermarks. Existing watermarking methods struggle to achieve fine-grained watermark embedding with a few images of specific concept and prevent removal of watermarks through personalized fine-tuning. Therefore, we introduce a novel concept-oriented watermarking framework that seamlessly embeds imperceptible watermarks into the concept of diffusion models. We introduce Fidelity-preserving Latent Watermarking (FLW) to generate latent watermarks based on image characteristics and the Adversarial Watermarking Modulation module to prevent "jailbreaking" via personalized finetuning. To enhance U-Net's efficiency in learning watermark patterns with limited samples, we propose Efficient Concept Watermark Finetuning, which alternates optimization of model parameters for both watermark embedding and concept learning. We conduct extensive experiments and ablation studies to verify our framework. Our code is available atthis https URL.

@article{lei2025_2411.11688,
  title={ Conceptwm: A Diffusion Model Watermark for Concept Protection },
  author={ Liangqi Lei and Keke Gai and Jing Yu and Liehuang Zhu and Qi Wu },
  journal={arXiv preprint arXiv:2411.11688},
  year={ 2025 }
}