Thomas Besnier

Øster Voldgade 3, Copenhagen, Denmark

Freshly graduated from CRIStAL at the University of Lille, in France. I am working on developping algorithms and new methods for 3D/4D deformation models of non-rigid shapes. My work was supervised by Prof. Mohamed Daoudi (IMT Nord-Europe) and Sylvain Arguillère (Laboratoire Painlevé, CNRS).

I am currently enrolled as a Postdoc at the University of Copenhagen, in the Pioneer Centre for AI under the supervision of Stefan Sommer.

news

Mar 26, 2026	Our paper PaNDaS: Learnable Deformation Modeling with Localized Control has been accepted to CVPR 2026 and will be presented in Denver in June!
Nov 25, 2025	Successfully defended my PhD entitled Reparametrization invariance in geometric deep learning: applications to human body and face motion synthesis ! - Manuscript, Slides
Sep 30, 2024	Attending ECCV in Milan to present ScanTalk: 3D Talking Heads from Unregistered Scans !

ongoing projects

Novo Nordisk Foundation 2024-2029

DIGIFABA

Decoding faba bean yield stability: From in-field image data to mechanistic models

DIGIFABA combines in-field and drone imagery with statistical, machine-learning, and mechanistic modelling to identify the traits behind stable faba bean yields. The project aims to strengthen breeding strategies for drought-resilient, climate-robust faba bean varieties and support more sustainable agricultural practice.

3D reconstruction shape analysis growth models yield stability

selected publications

Conference

2026

PaNDaS: Learnable Deformation Modeling with Localized Control

Thomas Besnier, Emery Pierson, Sylvain Arguillere, and 2 more authors

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

BibTeX Project PDF

@misc{besnier2024partialnonrigiddeformationsinterpolations,
  title = {PaNDaS: Learnable Deformation Modeling with Localized Control},
  author = {Besnier, Thomas and Pierson, Emery and Arguillere, Sylvain and Ovsjanikov, Maks and Daoudi, Mohamed},
  year = {2026},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  eprint = {2412.02306},
  archiveprefix = {arXiv},
  primaryclass = {cs.CV},
  url = {https://arxiv.org/abs/2412.02306},
  category = {conference-papers}
}

Conference

2024

ScanTalk: 3D Talking Heads from Unregistered Scans

Federico Nocentini, Thomas Besnier, Claudio Ferrari, and 3 more authors

In Proceedings of the IEEE/CVF European Conference on Computer Vision (ECCV), 2024

BibTeX Project PDF

@misc{nocentini2024scantalk3dtalkingheads,
  title = {ScanTalk: 3D Talking Heads from Unregistered Scans},
  author = {Nocentini, Federico and Besnier, Thomas and Ferrari, Claudio and Arguillere, Sylvain and Berretti, Stefano and Daoudi, Mohamed},
  year = {2024},
  booktitle = {Proceedings of the IEEE/CVF European Conference on Computer Vision (ECCV)},
  eprint = {2403.10942},
  archiveprefix = {arXiv},
  primaryclass = {cs.CV},
  category = {conference-papers}
}