Submitted by Andreas Rössler.

Submission data

Full nameXceptionNet Full Image c40 (FaceForensics++)
DescriptionSee FaceForensics++ paper.
Publication titleFaceForensics++: Learning to Detect Manipulated Facial Images
Publication authorsAndreas Rössler, Davide Cozzolino, Luisa Verdoliva, Christian Riess, Justus Thies, Matthias Nießner
Publication venueICCV 2019
Publication URLhttps://arxiv.org/pdf/1901.08971.pdf
Input Data TypesUses Full Image
Programming language(s)Python, Pytorch
HardwareGeForce 1080TI
Websitehttps://www.niessnerlab.org/
Source code or download URLhttps://github.com/ondyari/FaceForensics
Submission creation date29 Aug, 2019
Last edited29 Aug, 2019

Binary Classification v3

InfoDeepfakesFace2FaceFaceSwapNeuralTexturesPristineTotal
permissive0.7450.7590.7090.7330.5100.624