Skip to content
FaceOff Technologies

Deepfake Video Forensics

Drop a video here

MP4 · MOV · WebM · MKV, up to 200 MB

The eight modules start as soon as a file lands · 1 credit

0

Analyses run

3

Credits left

0

In history

Recent analyses0 items

Nothing analysed yet. Runs from this session are listed here.

CreditsFree

3 / 3

1 credit per analysis, no account. Nothing you submit is kept once the analysis returns.

Credit terms
Reading a verdict
  • Synthetic

    The weighted score crossed 50. AI-generated or manipulated signals found.

  • Authentic

    The weighted score stayed under 50. Nothing synthetic found.

A verdict comes with every module’s own score, so you can see which checks carried it.

The eight modulesWeights
  • Vision-language30%
  • Temporal motion15%
  • Frequency (DCT)14%
  • ViT artefacts12%
  • Physics & optics10%
  • Lip-sync8%
  • Zero-shot6%
  • GAN fingerprint5%
PlatformWeb · API

This takes one video. The platform takes a queue of them.

Same eight modules, submitted through the API and spread across GPU workers, with the report and the annotated video at the end.

Book a walkthrough

Preview of the output format. Module scores and the verdict are derived in the browser from the file’s digest, not measured by the analysis engine. The SHA-256 is computed from the real bytes.

One video here, a queue of them on the platform

This console takes one video at a time. DeepVideoGuard is the platform that runs the same eight modules across GPU workers, and hands back the report and the annotated video at the end.