yunqing-me/WatermarkDM
Code of the paper: A Recipe for Watermarking Diffusion Models
This tool helps content creators and businesses protect images generated by AI diffusion models from unauthorized use. It embeds an invisible 'watermark' (either a binary code or a hidden image-text pair) directly into the AI model during its training. The output is a diffusion model capable of producing watermarked images, allowing you to prove ownership or track origin.
155 stars. No commits in the last 6 months.
Use this if you are developing or fine-tuning AI diffusion models and need a robust way to embed undetectable watermarks into the generated images to protect your intellectual property.
Not ideal if you want to watermark existing images or need a visible watermark for immediate public notice.
Stars
155
Forks
11
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Nov 13, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/yunqing-me/WatermarkDM"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
OPTML-Group/Unlearn-Saliency
[ICLR24 (Spotlight)] "SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in...
Shilin-LU/VINE
[ICLR 2025] "Robust Watermarking Using Generative Priors Against Image Editing: From...
WindVChen/DiffAttack
An unrestricted attack based on diffusion models that can achieve both good transferability and...
koninik/DiffusionPen
Official PyTorch Implementation of "DiffusionPen: Towards Controlling the Style of Handwritten...
Wuyxin/DISC
(ICML 2023) Discover and Cure: Concept-aware Mitigation of Spurious Correlation