shivangi-aneja/TAFIM
[ECCV 2022] TAFIM: Targeted Adversarial Attacks against Facial Image Manipulation
This project helps individuals and organizations protect their facial images from unauthorized digital manipulation. It takes an original facial image and adds subtle, imperceptible changes. These 'protected' images, when fed into common face manipulation software, will instead produce a distorted, unusable output, effectively preventing deepfake creation or other alterations. This tool is designed for anyone concerned about image privacy and the misuse of their likeness through face manipulation technologies.
No commits in the last 6 months.
Use this if you want to prevent specific facial images from being altered by deepfake or other face manipulation models, ensuring they produce a garbled output instead of a convincing fake.
Not ideal if you are looking to detect manipulated images rather than prevent their creation, or if you need to protect images against forms of manipulation other than facial alterations.
Stars
58
Forks
3
Language
Python
License
—
Category
Last pushed
Dec 12, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/shivangi-aneja/TAFIM"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.