Hi! I am a researcher at IGN, the French National Geographical Institute, in the computer science department STRUDEL. The focus of my research is to exploit the structure of spatial/temporal data to improve the precision and speed of learning methods.
During my PhD I studied learning and optimization structured by large graphs (total variation, graphical models, etc). I now work on remote sensing application such as 3D point clouds and satellite hyperspectral imagery. I have also developed an interest for structured deep models, such as graph convolutions, recurrent neural network, and attention-based encoders.
See my academic_CV for more details (updated in Oct 2020).
NEW: Our paper presenting Torch-points3D havs been accepted as an oral at 3DV 2020. Congrats to the awesome dev team for their impressive work.
Leveraging Class Hierarchies with Metric-Guided Prototype Learning
Vivien Sainte Fare Garnot, Loic Landrieu, 2020 preprint
Torch-Points3D: A Modular Multi-Task Framework for Reproducible Deep Learning on 3D Point Clouds, Thomas Chaton, Nicolas Chaulet, Sofiane Horache, Loic Landrieu, 3DV2020 (oral) arxiv
Satellite Image Time Series Classification with Pixel-Set Encoders and Temporal Self-Attention, Vivien Sainte Fare Garnot, Loic Landrieu, Sebastien Giordano, Nesrine Chehata, CVPR2020 (oral). [arXiv]
Point Cloud Oversegmentation with Graph-Structured Deep Metric Learning, Loic Landrieu, Mohamed Boussaha, CVPR2019. [arXiv]
Cut-Pursuit Algorithm for Regularizing Nonsmooth Functionals with Graph Total Variation, Hugo Raguet and Loic Landrieu, ICML2018 [arXiv]
Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs, Loic landrieu and Martin Simonovsky, CVPR2018. [arXiv]
List of current and past interns:
Stephane Guinard, Simon Bailly, Omar Lahbib, Joana Roussillon, Thomas Luo (Helix.Re), Anna Kondracka (Vermessung AVT), Lamiae El-Mendili, Julien Baconat, Ameur Zaibi, Cédric Baron.
I teach machine learning at master level at ENSG and ENPC.
I review for ICML, NIPS, ICCV, CVPR, ACCV, BMVC, IJCV, PAMI, ANRT, ISPRS etc…