Julian Tachella

CNRS & ENS de Lyon

profile2.jpg

julian.tachella@cnrs.fr

I am a research scientist at the French National Centre for Scientific Research CNRS, working at the Sisyph laboratory, École Normale Supérieure de Lyon (Lyon, France). I am also the co-founder and CSO of Blur Labs.

My work lies at the intersection of signal processing and machine learning. I am particularly interested in the theory of imaging inverse problems and applications in computational imaging.

I am a lead developer of the deep inverse, an open-source library for solving inverse problems with deep learning.

Open PhD and master positions can be found here.

Code and videos related to my publications can be found here.

news

Oct 27, 2025 We are organizing a spring school on deep learning in Argentina, with participants from all over Latin America.
Oct 07, 2025 I have co-founded a startup building foundational models for computational imaging.
Mar 15, 2025 A 3-hour self-supervised tutorial is now available in YouTube! find videos and more here.
Feb 07, 2025 Unknown Noise Level Stein’s Unbiased Estimate is accepted at ICLR’25.
Jan 04, 2025 Our work on generalized recorrupted2recorrupted is accepted at CVPR’25.

selected publications

  1. terris2025ram.png
    Reconstruct Anything Model: a lightweight foundational model for computational imaging
    Matthieu Terris , Samuel Hurault , Maxime Song , and Julián Tachella
    2025
  2. UNSURE_demo.png
    UNSURE: self-supervised learning with Unknown Noise level and Stein’s Unbiased Risk Estimate
    Julian Tachella, Mike Davies , and Laurent Jacques
    ICLR 2025, 2025
  3. m_regimes_website.png
    Sensing Theorems for Unsupervised Learning in Linear Inverse Problems
    Julian Tachella, Dongdong Chen , and Mike Davies
    Journal of Machine Learning Research (JMLR), Jan 2023