Hi! I am a researcher at IGN, the French National Geographical Institute, in the machine learning team STRUDEL. The focus of my research is to develop new optimization and learning methods to exploit the structure of remote sensing data (spectral, spatial, temporal, multi-modal) for improved precision and speed.
During my PhD at INRIA/ENPC I studied learning and optimization structured by large graphs (total variation, graphical models, etc). I now develop machine learning and computer vision algorithms to address spatially-structured problems such as 3D point clouds or satellite images time-series analysis.
See my academic_CV for more details (updated in April 2021).
I am the main investigator of the ReADy3D ANR project on Real-time Analysis of Dynamic 3D data. I also am the co-program chair of the XXIVth ISPRS congress in Nice in 2022.
NEW: Our paper on panoptic segmentation of satellite time series, with Vivien Sainte-Fare Garnot has been accepted at ICCV 2021. Preprint.
NEW: Our paper on scalable deep surface reconstruction, with Raphael Sulzer, Renaud Marlet, and Bruno Vallet has been accepted at Eurographics SGP 2021. Preprint.
NEW: Our ISPRS Special Issue « Muti-modal learning in photogrammetry and remote sensing« , co-edited with Michael Ying Yang, Devis Tuia, and Charles Toth is online. Some very interesting work about a hot topic.
NEW: Our CVPR Workshop EarthVision have been accepted. Great line-up of speakers, contests, and as always a must for meeting remote-sensing researchers interested in the latest advances in computer vision.
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.
NEW: Our newest paper on satellite time series with Vivien Sainte-Fare Garnot has been acccepted at CVPR 2020 for an oral! Check the paper and the code. Comes with a new benchmark too.
Selected Articles
Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networks, Vivien Sainte Fare Garnot Loic Landrieu, ICCV2021 arXiv
Scalable Surface Reconstructionwith Delaunay-Graph Neural Networks, Raphael Sulzer, Loic Landrieu, Renaud Marlet, and Bruno Vallet, Eurographics SGP 2021 arXiv
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]
STUDENTS:
List of current and past PhD/Post Doc students:
Stéphane Guinard (Univ. Laval)
Mohamed Boussaha (Gambi-M)
Vivien Sainte-Fare Garnot
Raphael Sulzer
Romain Loiseau
Damien Robert
Ekaterina Kalinicheva (post doc)
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, Ameur Zaibi, Julien Baconat, Cédric Baron, Félix Quinton.
I teach machine learning at master level at ENSG and ENPC.
I am a technical advisor for SAMP, a startup using machine learning for producing digital twins of industrial facilities.
I review for ICML, NIPS, ICCV, CVPR (outstanding reviewer 2021), ACCV, BMVC, IJCV, PAMI, ANRT, ISPRS, TIP, etc…
