Research interests
My research activities mainly concern the study of inverse problems involving image data
Methodologies: plug and play, optimal transportation, variational methods, machine learning, data assimilation
Applications: medical imaging, computer vision, computational photography, oceanography
Resume
I received the M. Eng degree in applied mathematics from the National Institute of Applied Sciences, Rouen, France, in 2004 and the M. Sc in Analysis from Rouen University in 2004. I obtained the Ph.D. degree in applied mathematics from the University of Rennes, France, in 2007, under the supervision of Étienne Mémin in the Inria team Vista at IRISA, with long stays at the Laboratorio de Fluiodinámica de Buenos Aires, Argentina in 2006 and 2007. From 2008 to 2010, I had a postdoctoral position in Barcelona Media, Spain, under the supervision of Vicent Caselles. I joined the CNRS (section 07) as a full researcher in 2010 in the Inria team MOISE (now AIRSEA team) at the Laboratoire Jean Kuntzmann, Grenoble, France. Since 2013, I work in the Institut de Mathématiques de Bordeaux, affiliated to IOP and CSM teams. I realized long stays in the Cambridge Image Analysis group at the Department of Applied Mathematics and Theoretical Physics of the University of Cambridge, UK, in 2018-2019. Since 2023, I am a member of the Inria team MONC.
PhD supervision
- Valentine Tosel (2025 - , Univ. Bordeaux, co-dir with J. Delon): Regularization of inverse problems with generative models
- Antonin Clerc (2025 - , Univ. Bordeaux, co-dir with B. Denis de Senneville): Plug-and-Play methods for MRI
- Guido Tapia Riera (2025 - , Univ. Bordeaux, co-dir with C. Castera): Minmax optimization
- Idris Tatachak (2025 -, Univ. Lyon, Acteon, co-dir with S. Rit and X. Ripoche): Dental CT reconstruction
- Marien Renaud (2023 - 2026, Univ. Bordeaux, co-dir with A. Leclaire): Stochastic Plug-and-Play methods for radar image restoration
- Samuel Hurault (2020 - 2023, Univ. Bordeaux, co-dir with A. Leclaire): Convergent plug-and-play methods for image inverse problems with explicit and nonconvex deep regularization
- Jean Prost (2020 - 2023, Univ. Bordeaux, co-dir with A. Almansa): Image restoration with deep generative models
- Warren Jouanneau (2020 - 2023, Univ. Bordeaux, Lectra, co-dir with A. Bugeau, M. Palyart and L. Vézard): Segmentation and weakly-supervized classification of fashion images
- Elsa Cazelles (2015 - 2018, Univ. Bordeaux, co-dir with J. Bigot): Statistical aspects of Wasserstein barycenters
- Rémi Giraud (2014 - 2017, Univ. Bordeaux, co-dir with V.-T. Ta): Superpixel and Non-Local methods
- Romain Hug (2012 - 2016, Univ. Grenoble Alpes, co-dir with E. Maitre): Numerical Analysis of Dynamical Optimal transport
- Romain Yildizoglu (2012 - 2014, Univ. Bordeaux, co-dir with J.-F. Aujol and C. Dossal): Convexification methods for segmentation
PostDocs
- Antoine Houdard (2019 - 2021, Univ. Bordeaux, co-dir with J. Rabin and A. Leclaire): Machine Learning and Optimal Transport
- Arnaud Dessein (2015 - 2016, Univ. Bordeaux,co-dir with J.-L. Rouas): Optimal Transport for Audio processing
- Alexandros Makris (2011 - 2012, Univ. Joseph Fourier): Sequential Image Assimilation
Engineer
- Julie Lesthelle (2026, Univ. Bordeaux, co-supervision with C. Etchegaray and S. Duchesne): Les études en licence de mathématiques sous le prisme du genre : Undergraduate Mathematics Studies through a Gender Lens: Constructing a Masculine Environment and the Experiences of Women as a Minority
Recent responsabilities
- Head of the Inria Project-Team MONC (since sept. 2024)
- Deputy director of the Institut de Mathématiques de Bordeaux, in charge of societal challenges (since 2023)
- Direction team of the Research Network MAIAGES - MAthématiques de l’Imagerie, Apprentissage et GEométrie Stochastique (2024-205)

- Direction, with Julie Delon, of the Research Network MIA - Mathématiques de l’Imagerie et de ses Applications (2022-2023)
- Member of the French National Research Agency committee for Artificial Intelligence (2020-2022)
Last projects
- RALMI (2026) : Recherche Action pour une Licence de Mathématiques plus Inclusive
- ANR OPLA (2026 - 2030): Optimal MR Protocol for the monitoring of SVD patients at Low mAgnetic field
- BPI ConeBEAM AI (2025 - 2029): dental CBCT recontruction
- PEPR PDE AI (2023 - 2028): Partial Differential Equations for Artificial Intelligence: numerical analysis, optimal control and optimal transport
- ANR Holibrain (2023 - 2028): Holistic Brain Analysis
- ANR PostProdLEAP (2019- 2024): rethinking archive PostProduction with Learning, vAriational, and Patch-based methods
- H2020 NoMADS (2018 - 2024): Nonlocal Methods for Arbitrary Data Sources
- ANR GOTMI (2016 - 2021): Generalized Optimal Transport Models for Image processing
Organization of scientific events
- Neural Information Processing Systems (NeurIPS'26), Paris, 2026
- Workshop on Optimization and Sampling for Inverse Problems, Bordeaux, 2026
- International Conference on Mathematical Image Analysis, Paris, 2025
- International Conference on Mathematical Image Analysis, Berlin, 2023
- French-Italian workshop on Mathematics of Imaging, Vision and their Applications, Nice, 2022
- International Conference on Mathematical Image Analysis, virtual, 2021
Recent talks
- EDP et Calcul Scientifique seminar), LMRS, Rouen, 2026
- ICMS workshop on Imaging inverse problems and generating models, Edimbourgh, 2026
- Imaging in Paris seminar, 2026
- Workshop on (blind) inverse problems in imaging, CIRM, Marseille, 2025
- Image seminar, GREYC, Cane, 2025
- Oberwolfach workshop on Mathematical Imaging and Surface Processing, Allemagne, 2025
- Séminaire Inria, Bordeaux, 2024
- Séminaire Cambridge Image Analysis, Royaume-Uni, 2023
- Journées de Statistique de la SFdS, Lyon (canceled for Covid), 2022
- Mathematics of Deep Learning from Universität Erlangen-Nürnberg, virtuel 2021
- Workshop on AI for Ocean, Atmosphere and Climate Dynamics, Brest, 2020