Mamadou-Keita/VLM-DETECT
[ICASSP 2024] The official repo for Harnessing the Power of Large Vision Language Models for Synthetic Image Detection
This project helps identify whether an image is real or was generated by an AI model, often called a 'deepfake.' You provide it with images, and it tells you if they are synthetic. This is useful for content moderators, journalists, and forensic analysts who need to verify image authenticity.
No commits in the last 6 months.
Use this if you need to reliably determine if an image was created by a diffusion-based AI model (like Stable Diffusion or DALL-E) or is a genuine photograph.
Not ideal if you're looking to detect manipulated images that started as real photos or if you need to identify the specific AI model used for generation.
Stars
34
Forks
2
Language
Python
License
—
Category
Last pushed
Aug 13, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Mamadou-Keita/VLM-DETECT"
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