ondyari/FaceForensics resource

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

2764 stars · 590 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

FaceForensics++ is a forensics benchmark dataset of 1000 original YouTube video sequences manipulated with four automated face manipulation methods (Deepfakes, Face2Face, FaceSwap, NeuralTextures), plus the Google/Jigsaw Deep Fake Detection dataset. It includes binary masks and 1000 Deepfakes generation models for training image/video forgery detection and segmentation models.

Use cases

• train a deepfake detection model
• benchmark face forgery detection methods
• get labeled real and manipulated face videos for research
• train a segmentation model to localize manipulated face regions
• generate augmented deepfake training data with provided models
• study face reenactment and face swapping techniques

When to choose

• you need a widely used academic benchmark for facial forgery detection
• you want real and fake video pairs with pixel-level masks
• you need diverse manipulation methods (deepfakes, face2face, faceswap, neuraltextures) in one dataset

When to avoid

• you need data without agreeing to the dataset's terms of use and access request process
• you need high-resolution or unconstrained faces with occlusions
• you need a commercially licensed dataset for product use

Facets

dataset · maturity stable

machine-learning computer-vision image-processing video-processing data-science computer-vision deep-learning artificial-intelligence security python cross-platform deepfake-detection face-forensics video-dataset face-manipulation benchmark face-swapping segmentation-masks

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ondyari/FaceForensics")