Samuel Hurault

Samuel Hurault

Postdoctoral researcher

I am a postdoctoral researcher at ENS Paris working with with Gabriel Peyré. I obtained my Ph.D from Université de Bordeaux, co-supervised by Nicolas Papadakis and Arthur Leclaire. My research interests include image restoration, generative modeling, nonconvex optimization. The main objective of my PhD was to develop inovative deep denoising priors along with convergent optimization schemes for solving image inverse problems. I am also one of the main developer of the Python library DeepInv, which provides a unified framework for solving inverse problems using deep neural networks.

Download my resume.

Interests
  • Optimization
  • Computer Vision
  • Machine Learning
  • Signal Processing
Education
  • PhD in Applied Mathematics, 2023

    Université of Bordeaux, France

  • MRes MVA (Mathematiques, Vision, Apprentissage), 2019

    Université Paris-Saclay, France

  • MRes in Applied Mathematics, 2018

    Ecole Normale Superieure Paris-Saclay, France

  • BSc in Mathematics, 2017

    Ecole Normale Superieure Paris-Saclay, France

Publications

(2023). Convergent Bregman Plug-and-Play Image Restoration for Poisson Inverse Problems. In Neurips 2023.

PDF Cite

(2023). A relaxed proximal gradient descent algorithm for convergent plug-and-play with proximal denoiser. In SSVM 2023.

PDF Cite

(2023). Self-Consistent Velocity Matching of Probability Flows. In Neurips 2023.

PDF Cite

(2022). An Analysis of Generative Methods for Multiple Image Inpainting. Handbook of Mathematical Models and Algorithms in Computer Vision and Imaging.

PDF Cite

(2022). Proximal denoiser for convergent plug-and-play optimization with nonconvex regularization. In ICML 2022.

PDF Cite Code Video

(2021). Gradient Step Denoiser for convergent Plug-and-Play. In ICLR 2021.

PDF Cite Code Video

(2020). Self-Supervised Small Soccer Player Detection and Tracking. ACM MMSports 2020.

PDF Cite Code

Talks

Contact