Thomas Eboli

Postdoctoral researcher

Centre Borelli, ENS Paris-Saclay

Office 2S26
firstname dot lastname at ens-paris-saclay dot fr
Google scholar / GitHub / LinkedIn / YouTube

Bio

I am a postdoctoral researcher at ENS Paris-Saclay under the surpervision of Jean-Michel Morel and Gabriele Facciolo. Before that, I was a PhD student at INRIA Paris, in the WILLOW project team under the supervision of Jean Ponce and Jian Sun. I worked on hybrid methods for image deblurring, at the intersection of classical inverse problems and recent deep learning techniques. Prior to research, I received the BS and the MS degrees from Telecom Paris and the MVA MS degree from ENS Paris-Saclay. During my PhD, I spent time at NYU Center for Data Science. My research focuses on computational photography and image processing, with a particular interest on image and video deblurring and super-resolution algorithms and their practical applications.


News

Activities


Publications

Collaborative Blind Image Deblurring

Thomas Eboli, Jean-Michel Morel, Gabriele Facciolo
Arxiv, 2023
[paper] [code] [webpage] [bibtex]

SING: A Plug-and-Play DNN Learning Technique

Adrien Courtois, Damien Scieur, Jean-Michel Morel, Pablo Arias, Thomas Eboli
Arxiv, 2023
[paper] [code] [bibtex]

Realistic Low-Resolution Raw Image Simulation for Learning Super-Resolution
Realistic Low-Resolution Raw Image Simulation for Learning Super-Resolution

Thomas Eboli, Jamy Lafenetre, Jean-Michel Morel, Gabriele Facciolo
Arxiv, 2023
[paper]

Fast Chromatic Aberration Correction with 1D Filters

Thomas Eboli
IPOL, 2023
[paper] [code] [demo] [bibtex]

Handheld Burst Super-Resolution Meets Multi-Exposure Satellite Imagery

Jamy Lafenetre, Ngoc Long Nguyen, Gabriele Facciolo, Thomas Eboli
CVPRW, 2023
[paper] [code] [poster] [bibtex]

Implementing Handheld Burst Super-Resolution

Jamy Lafenetre, Gabriele Facciolo, Thomas Eboli
IPOL, 2023
[paper] [code] [demo] [bibtex]

Breaking down Polyblur: Fast Blind Correction of Small Anisotropic Blurs

Thomas Eboli, Jean-Michel Morel, Gabriele Facciolo
IPOL, 2022
[paper] [code] [pypi] [demo] [bibtex]

Fast Two-step Blind Optical Aberration Correction

Thomas Eboli, Jean-Michel Morel, Gabriele Facciolo
ECCV, 2022
[paper] [code] [webpage] [demo] [video] [poster] [bibtex]

High Dynamic Range and Super-Resolution from Raw Image Bursts

Bruno Lecouat, Thomas Eboli, Jean Ponce, Julien Mairal
SIGGRAPH, 2022
[paper] [webpage] [video] [bibtex]

Learning to Jointly Deblur, Demosaick and Denoise Raw Images

Thomas Eboli, Jian Sun, Jean Ponce
Arxiv, 2021
[paper]

Structured and Localized Image Restoration

Thomas Eboli*, Alex Nowak Vila*, Jian Sun, Francis Bach, Jean Ponce, Alessandro Rudi
Arxiv, 2020
[paper]

End-to-End Interpretable Learning of Non-Blind Image Deblurring

Thomas Eboli, Jian Sun, Jean Ponce
ECCV, 2020
[paper] [code] [bibtex]

Open source

Open-source research is important, and I strive (with the help of others when possible) to reproduce important papers's missing implementations and release the codes on Github.

People

Awards

Teaching

Talks

Computational photography awesome resources

Here are some resources I really like to better understanding principles from many fields related to computer vision, and in particular computational photograpy and image processing. I often relate to them as off-the-road textbooks. Kudos to the authors!

Miscellaneous

I am fond of soccer (go PSG), running, cycling, opera and photography (proud owner of a Sony A6600 camera). I am always OK to chat about these topics around a cup of coffee!