Shilin-LU/VINE
[ICLR 2025] "Robust Watermarking Using Generative Priors Against Image Editing: From Benchmarking to Advances" (Official Implementation)
This project helps artists, photographers, and content creators protect their digital images from unauthorized use, especially when those images might be heavily edited or transformed by AI tools. You input your original image and a secret message, and it outputs a watermarked image that looks nearly identical to your original but secretly contains your hidden message. This is designed for anyone creating visual content who needs to ensure ownership even after extensive modifications.
385 stars.
Use this if you need to embed a robust, invisible watermark into your images that can survive various edits like regeneration, global transformations, or local retouching, ensuring you can still prove ownership.
Not ideal if you need a visible watermark, or if your primary concern is preventing access to the original image rather than proving its origin after modification.
Stars
385
Forks
37
Language
Python
License
—
Category
Last pushed
Dec 01, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Shilin-LU/VINE"
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...
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
bytedance/LatentUnfold
Implementation of paper: Flux Already Knows – Activating Subject-Driven Image Generation without Training