davide-coccomini/MINTIME-Multi-Identity-size-iNvariant-TIMEsformer-for-Video-Deepfake-Detection

Code for Video Deepfake Detector from "MINTIME: Multi-Identity Size-Invariant Video Deepfake Detection", paper available on IEEE Transactions on Information Forensics and Security.

34
/ 100
Emerging

This project helps identify manipulated videos, known as deepfakes, by analyzing visual information to determine if a video is authentic. You provide a video file as input, and it tells you if the video contains deepfakes, and can even highlight which individuals in multi-person videos might be manipulated. It's designed for anyone needing to verify video authenticity, such as journalists, content moderators, or legal professionals.

No commits in the last 6 months.

Use this if you need to reliably detect manipulated videos, especially those with multiple people or varying face sizes, and understand where the manipulation might be occurring.

Not ideal if you need to analyze static images for manipulation, as this tool focuses specifically on temporal and spatial anomalies within video.

video-verification digital-forensics content-moderation media-authenticity
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 18 / 25

How are scores calculated?

Stars

68

Forks

13

Language

Jupyter Notebook

License

Last pushed

Jun 05, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/davide-coccomini/MINTIME-Multi-Identity-size-iNvariant-TIMEsformer-for-Video-Deepfake-Detection"

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