taco-group/HFMF

HFMF: Hierarchical Fusion Meets Multi-Stream Models for Deepfake Detection

24
/ 100
Experimental

This project offers a sophisticated tool for detecting deepfakes in images and videos. It takes visual media as input and outputs a clear indication of whether the content is a deepfake, along with visual heatmaps highlighting manipulated regions. Digital forensic investigators, content moderators, and journalists can use this to verify media authenticity.

No commits in the last 6 months.

Use this if you need to reliably identify highly realistic deepfakes generated by advanced AI, and also understand why the detection was made.

Not ideal if you require a simple, ready-to-use plug-and-play solution without any setup involving code or model weights.

deepfake-detection media-authenticity digital-forensics content-moderation fake-news-analysis
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 11 / 25

How are scores calculated?

Stars

13

Forks

2

Language

Jupyter Notebook

License

Last pushed

Jan 06, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/taco-group/HFMF"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.