lara raad cisa



about me

I am an associate professor (Maîtresse de Conférences) at ESIEE Paris, France.

Previously, I spent one year as a Postdoc reseacher at Télécom Paris, France, with Yann Gousseau and Bruno Galerne and three 
years at Universitat Pompeu Fabra, Spain, with Coloma Ballester and Gloria Haro. Before that, I obtained my Ph.D. 
from ENS Paris Saclay, France, under the supervision of Agnès Desolneux and Jean-Michel Morel.

My research interests are in image processing and computer vision. Currently, I am mostly interested in the problem
of stochastic modelling of natural images and in particular exploring deep generative models and finding fundamental improvement
to these models by solving inverse problems.

publications

Influence of Color Spaces for Deep Learning Image Colorization
  A. Bugeau, R. Giraud, L. Raad
  Springer Handbook of Mathematical Models and Algorithms in Computer Vision and Imaging, 2022
  arXiv

Analysis of Different Losses for Deep Learning Image Colorization
  C. Ballester, A. Bugeau, H. Carrillo, M. Clément, R. Giraud, L. Raad, P. Vitoria
  arXiv preprint arXiv:2204.02980, 2022
  arXiv

On anisotropic optical flow inpainting algorithms
  L. Raad, M. Oliver, C. Ballester, G. Haro, E. Meinhardt
  Image Processing On Line, 2020
  doi web demo code

ChromaGAN: Adversarial Picture Colorization with Semantic Class Distribution
  P. Vitoria, L. Raad, C. Ballester
  Winter Conference on Applications of Computer Vision, 2020
  pdf

Motion Inpainting by an Image-Based Geodesic AMLE Method
  M. Oliver, L.Raad, C. Ballester, G. Haro
  IEEE International Conference on Image Processing, 2018
  pdf

A survey of exemplar-based texture synthesis
  L.Raad, A. Davy, A. Desolneux, J.-M. Morel
  Annals of Mathematical Sciences and Applications, 2018
  pdf

Efros and Freeman Image Quilting Algorithm for Texture Synthesis
  L.Raad and B. Galerne
  Image Processing On Line, 2017
  doi web demo code

Conditional and multiscale locally Gaussian models for texture synthesis
  L.Raad, A. Desolneux, J.-M. Morel
  Journal of Mathematical Imaging and Vision, 2016
  pdf

Multiscale Exemplar Based Texture Synthesis by Locally Gaussian Models
  L.Raad, A. Desolneux, J.-M. Morel
  Iberoamerican Congress on Pattern Recognition, 2015
  pdf

Conditional Gaussian Models for Texture Synthesis
  L.Raad, A. Desolneux, J.-M. Morel
  International Conference on Scale Space and Variational Methods in Computer Vision, 2015
  pdf

Locally Gaussian Exemplar Based Texture Synthesis
  L.Raad, A. Desolneux, J.-M. Morel
  IEEE International Conference on Image Processing, 2014
  pdf

patents

Surface covering production method using digital printing
  C. Maertens, R. Peres, L. Raad
  WO2017121825A1 (Application, January 2016)
  link

teaching

2021-2022 - ESIEE Paris

  Optimization algorithms
  C Progamming
  Algorithmics
  Algorithmics and Python programming

2017-2020 - Universitat Pompeu Fabra

  Image processing
  Computer Vision
  Numerical methods

2012-2016 - ENS Paris-Saclay

  Hilbert and Fourier analysis
  Numerical methods and ordinary differential equations
  Optimisation methods
  Introduction to GNU/Linux and LaTex