MesoNet: a Compact Facial Video Forgery Detection Network

MesoNet is a deep learning program dedicated to to automatically and efficiently detect face tampering in videos, and particularly focuses on two recent techniques used to generate hyper-realistic forged videos: Deepfake and Face2Face.

MesoNet: a Compact Facial Video Forgery Detection Network, Darius Afchar, Vincent Nozick, Junichi Yamagishi and Isao Echizen, in IEEE Workshop on Information Forensics and Security, WIFS, December 2018.

• paper: MesoNet_WIFS_2018.pdf on arxiv
• source code: MesoNet (git)

Computer graphics vs. photgraphic images

This paper presents a method to distinguish computer graphics from real photographic images. The program is based on a convolution neural network focussing on the statistical properties of the images, including image noise, to efficiently classify the data.

Distinguishing Computer Graphics from Natural Images Using Convolution Neural Networks, Nicolas Rahmouni, Vincent Nozick, Junichi Yamagishi and Isao Echizen, in IEEE Workshop on Information Forensics and Security, WIFS 2017, Rennes, France, December 2017.

• pdf: Rahmouni_WIFS_2017.pdf
• source code: CGvsPhoto (git)
• dataset: GameCG.zip