About me

I am an associate professor (Maitre de conférences) in signal processing, image processing and machine learning at Toulouse-INP - ENSEEIHT. I am also part of the Signal and Communication (SC) team at the Institut de Recherche en Informatique de Toulouse (IRIT).

In 2020, I graduated from INSA Rouen with an engeenering degree in applied mathematics, and I have obtained my PhD from Bordeaux university in 2023. My thesis title was Image restoration with deep generative models.

I have worked as a post-doctoral researcher at the MAP5 lab of université Paris-cité, and at the research center on computer science, signal processing and automatics of Lille (CRIStAL).

My main research goal is to conceive efficient and reliable algorithms for solving challenging inverse problems, in applications such as image processing or cosmology. To this end, I develop hybrid methods which incorporate deep neural networks within optimization or sampling algorithms in order to benefit both from the well-established mathematical guarantees of model-based approaches and the ability of deep neural networks to model complex data distributions.

Internship offer

We are proposing a master 2 internship on estimation of the covariance matrix for imaging inverse problem. This is a 6 months internship which will take place at IRIT and begin around february 2026. Please refer to the full description for more information.

Publications and preprints

• Un réseau de neurones augmenté pour simuler l'évolution de la distribution de matière noire
  Jean Prost, Pierre-Antoine Thouvenin, Jenny Sorce, Pierre Chainais
  (GRETSI2025) [arxiv]
• LATINO-PRO: LAtent consisTency INverse sOlver with PRompt Optimization
  Alessio Spagnoletti, Jean Prost, Andrés Almansa, Nicolas Papadakis, Marcelo Pereyra
  (ICCV2025) [arxiv] [paper] [code]
• Efficient posterior sampling for diverse super-resolution with hierarchical VAE Prior
  Jean Prost, Antoine Houdard, Nicolas Papadakis, Andrés Almansa
  19th International Joint Conference on Computer Vision Theory and Applications (VISAPP2024) [arxiv]
• Plug-and-Play image restoration with Stochastic deNOising REgularization
  Marien Renaud, Jean Prost, Arthur Leclaire, Nicolas Papadakis
  Forty-first International Conference on Machine Learning [arxiv] [paper] [code]
• Inverse problem regularization with hierarchical variational autoencoders
  Jean Prost, Antoine Houdard, Nicolas Papadakis, Andrés Almansa
  International Conference on Computer Vision (ICCV 2023) [arxiv] [paper] [code]
• SCOTCH and SODA: A Transformer Video Shadow Detection Framework
  Lihao Liu, Jean Prost, Lei Zhu, Nicolas Papadakis, Pietro Lio, Carola-Bibiane Schönlieb, Angelica I Aviles-Rivero
  Computer Vision and Pattern Recognition (CVPR 2023) [arxiv] [paper] [code]
• Learning Local Regularization for Variational Image Restoration
  Jean Prost, Antoine Houdard, Andrés Almansa, Nicolas Papadakis
  Scale Space and Variational Methods in Computer Vision: 8th International Conference, (SSVM 2021) [arxiv] [paper] [code]