Vincent Nozick

Vincent Nozick

Associate Professor at Université Gustave Eiffel.
Laboratoire d'informatique Gaspard Monge (LIGM).
In charge of Master 2 sciences de l'image.


email :
vincent.nozick (a)

address :
Laboratoire d'informatique Gaspard Monge
Université Gustave Eiffel
Cite DESCARTES, 5 boulevard Descartes
77454 Marne-la-Vallée CEDEX 2

phone number :
ESIEE (office 5357): 01 45 92 67 03


  • Geometric algebra
  • Computer vision
  • Digital image forensics

Professional background

  • 2019 : Habilitation a diriger les recherches (HDR)
  • since 2008: Maître de conferences (associate professor) : Université Gustave Eiffel, LIGM-A3SI, France
  • 2016-2018: sabbatical at JFLI, Tokyo, Japan
  • 2016-2018: invited researcher at Hideo Saito Lab, Keio University, Japan
  • 2011-2013: Headmaster of IMAC engineer school
  • 2007-2008: Research Associate : Keio University, Japan
  • 2006-2007: Post-doc : Hideo Saito lab, Keio University, Japan
  • 2002-2006: PhD degree in computer sciences


Under (re)construction ... See my google scholar instead.

Geometric algebra and Garamon

Garamon stands for Geometric Algebra Recursive and Adaptative Monster. It is a C++ library generator synthesizing efficient C++ libraries implementing geometric algebras in both low and higher dimensions, with any arbitrary metric. The library generator is designed to produce easy to install, easy to use, effective and numerically stable libraries. The design of the libraries is based on a prefix tree data structure and a recursive scheme for high dimensions.

The source code is available online:

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.

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.

Multiple images rectification

  • Multiple image rectification in C++ and in matlab.





Pres on the web

Written press



  • National Institute of Informatics : NII
  • Laboratoire d'Informatique Gaspard Monge (UPEM) : LIGM
  • CNRS : INS2I
  • Université Paris-Est Marne-la-Vallée : UPEM