Publications

  • [GPC*16] Gafar, A., Perrier, H., Coeurjolly, D., Ostromoukhov, V., Guo, J., Yan, D., Huang, H., Deussen, O. Low Discrepancy Blue Noise Sampling. In ACM Transactions on Graphics (Proceedings of ACM SIGGRAPH Asia 2016).
    Web page for the projet

  • [PCL*16] Perrier, H., Levallois, J., Coeurjolly, D., Farrugia, J.-P., Iehl, J.-C., Lachaud, J.-O., Interactive Curvature Tensor Visualization on Digital Surfaces. Discrete Geometry for Computer Imagery : 19th IAPR International Conference, DGCI 2016, Nantes, France, April 18-20, 2016.
    GIT repository for the projet

Experience

  • October 2014 - now
    These - LIRIS, Team R3am
    Supervised by - Victor Ostromoukhov, David Coeurjolly
    To display a 3D scene on a computer, we need to be able to convert this object into a 2D RGB image, wich means finding a proper color for each pixel of this image. To find this color, we integrate all the light that bounces back to this pixel after propagating in the scene. We can solve this integration by finding an anlytical expression for it, or by doing numerical integration with sampling based method (such as Monte Carlo). My PhD focused on the various existing sampling algorithms and their properties in order to optimise this integration.
  • February 2014 - September 2014 - Master's Internship
    Continuous génération of Level of Détails - LIRIS, Team R3am
    Supervised by - Jean Claude Iehl, Jean Phillipe Farrugia
    Our goal was to find an efficient GPU implementation of an adaptive remeshing for isosurface extraction. By using the parallism of the GPU and their ability to efficiently process triangles, we were able to build the mesh and deal with the level of details entirely on the GPU and in a very short time.
    OpenGL 3+, C++, Git
    Prensented during the "Journée de l'AFIG 2014", slides here
  • July 2013 - August 2013 - Internship
    Multi-scale statistical representation of point clouds - LIRIS, Team R3am
    We wanted to define an efficient representation for rendering 3D scenes defined with point clouds. Our representation was supposed to have at the same time a low memory footprint and to allow efficient rendering through multi scaling. To do so, we used a Bounding Volume Hiearchy (BVH, a binary tree) and stored whithing each node a gaussian function representing the distribution of the points in this volume.
    OpenGL 3+, C++, Git
  • May 2012 - August 2012 - Bachelor's degree Internship
    Adaptation and study of 3D reconstruction algorithm - Awabot SAS
    The issue is to reconstruct in real time the environnement in which a robot was evolving by using a RGB and Depth map. My work was to port the RGBDSLAM algorithm, written within the ROS framework, into the framework of the robot Emox. I also studied the PCL implementation of the algorithm Kinect Fusion to adapt it to a Time of Fly camera.
    PCL, OpenCV, OpenGL, C++
  • April 2011 - August 2011 - DUT's Internship
    Creation of augmented reality based video games for the robot Emox - Awabot SAS
    I developped a simple game of "Shoot them up" within the framework of the robot Emox. The game is entirely in augmented reality using the robot's camera and is controlled by markers within the environnement.
    C++, Urbi

Education

  • October 2014 - now
    Phd en Computer graphics, LIRIS Laboratory
    University Lyon 1
  • 2012 - 2014
    Master in Computer sciences, oriented Computer Graphics
    University Lyon 1, Top 5%
  • 2011 - 2012
    Bachelor (last year) in Computer Sciences
    University Lyon 1, Top 5%
  • 2009 - 2011
    DUT (equivalent to the first two years of Bachelor) in Computer Sciences
    IUT A Lyon 1, Top 10%

Languages

  • English, read written spoken (970 TOEIC)
  • Français, mothertongue
  • Spanish, beginner
  • Russian, beginner
top