Education
INRIA, University of Rennes
Ph. D in 3D Computer Vision, advised by Adnane Boukhayma and Eric Marchand.
Ph. D in 3D Computer Vision, advised by Adnane Boukhayma and Eric Marchand.
Dec 2022 - Dec 2025
IMT Atlantique
Master Degree in Computational and Mathematical Engineering.
Master Degree in Computational and Mathematical Engineering.
Sep 2018 - Aug 2022
Lycée du Parc
Preparatory classes for French engineering schools.
Preparatory classes for French engineering schools.
Sep 2017 - Aug 2018
Employment
Inria
Research Engineer working on improving the generalization 3D shape reconstruction methods.
Research Engineer working on improving the generalization 3D shape reconstruction methods.
Apr 2021 - Sep 2021
Rennes, France
Rennes, France
OSO-AI
Research Engineer working on Graph-based learning of acoustic representations.
Research Engineer working on Graph-based learning of acoustic representations.
Apr 2020 - Feb 2021
Brest, France
Brest, France
Maroc Bureau
Software Engineer working on Database management.
Software Engineer working on Database management.
Apr 2019 - Jun 2019
Brest, France
Brest, France
Publications
SparseCraft: Few-Shot Neural Reconstruction through Stereopsis Guided Geometric Linearization
Imporving Novel-View Synthesis and reconstruction from sparse views without data priors through local Geometric Linearization.
ECCV, 2024
Unsupervised Occupancy Learning from Sparse Point Cloud
Local linearization and Minimal entropy fields regularization for occupancy learning from sparse point cloud without ground-truth supervision.
CVPR, 2024
Few-Shot Unsupervised Implicit Neural Shape Representation Learning with Spatial Adversaries
Adversarial regularization for SDF optimization from sparse point cloud without ground-truth supervision.
ICML, 2024
Mixing-denoising generalizable occupancy networks
Making generalizable implicit representations robust to the input noise and sparsity by relaxing the inductive biais of convolutions with MLP-based local encoding.
3DV, 2024
Robustifying Generalizable Implicit Shape Networks with a Tunable Non-Parametric Model
Leveraging kernel methods for Fast test-time adaptation of pretrained networks with a shape specific trade-of between expressiveness and regularization.
NeurIPS, 2024
Few ‘Zero Level Set’-Shot Learning of Shape Signed Distance Functions in Feature Space
Conditioning implicit representations on sparse inputs through meta-leanring in feature space.
ECCV, 2022