SenRVM : série temporelle de NDVI sans lacunes par régression multimodale pour le suivi de la végétation cesbio.cnrs.fr/senrvm-serie-t… "Recurrent-based regression of Sentinel time series for continuous vegetation monitoring" Remote Sensing of Environment hal.archives-ouvertes.fr/hal-03274870/

The month of May was somewhat hectic but rich in emotions. ..successfully defended my PhD ! 🤘🎓🌎🛰️ADEME CNES IGN France Centre d'Etudes Spatiales de la Biosphère ..attended the amazing European Space Agency Living Planet Symposium (#LPS22) ! ..got a new job at IGN France! Now, for the lucky ones, see you in Nice ISPRS🎊!

Launch of #IGN 1st #IA land cover challenge: #FLAIR-one on codalab.org. Sign up to use our FLAIR dataset (French Land cover from Aerial ImageRy), train #IA models and report your results during the next 4 months. Can you beat the benchmark!? Sign up now!

🚨Only 20 days left to participate in the challenge #IGN #AI land cover challenge: #FLAIR-one. Use our FLAIR dataset (French Land cover from Aerospace ImageRy), train #IA models and report your results. Almost 250 registered and more than 800 submissions 🙌codalab.lisn.upsaclay.fr/competitions/8…

🚨🚀✅🇫🇷 Happy to announce that our "GeoMultiTaskNet: remote sensing unsupervised domain adaption using geographical coordinates" was accepted at #CVPR2026 workshop EarthVision :) Thanks to Nicolas Gonthier, Clement Mallet, Garioud Anatol and Simone Scardapane! #earthvision #CVPR2023 1/2

Dataset discovery: The FLAIR-one semantic segmentation dataset consists of 77,412 high resolution patches (512x512 at 0.2 m spatial resolution) with 19 semantic classes ignf.github.io/FLAIR/

#CommunNumérique | A vous de jouer ! Après un 1er défi, #Flair est maintenant disponible en #Opensource sur #Github Ce jeu de données géospatiales décrit la diversité du territoire pour cartographier l’occupation du sol. Un #dataset incontournable pour entraîner des modèles #IA

#Digitalcommons | It's up to you! After a first challenge, #Flair is now available as #Opensource on #Github This geospatial dataset describes the diversity of the territory to map land cover. An essential resource to train #IA model 🤖🗺️ ℹ️ ignf.github.io/FLAIR/

[#ChallengeIA] Relevez le défi #FLAIR 2 ! Avec Connect by CNES et l’@ENSG 🔬+🛰️+🛩️+🤖 = 🗺️ Avec ce nouveau challenge #IA sur l’occupation des sols #OCSGE, le #Dataset s’enrichit de l’imagerie satellitaire du programme Copernicus EU #FPCUP Pour participer ⬇️

[#AIChallenge] Join the #FLAIR 2 challenge! With Connect by CNES and ENSG-Géomatique 🔬+🛰️+🛩️+🤖 = 🗺️ With this new #AI challenge on land cover, the #dataset is enriched with #satelliteimagery from the Copernicus EU program #FPCUP Check out ⬇️

[📣 Webinaire] On vous présente le projet #CoSIA, (Couverture du Sol par Intelligence Artificielle), ses données, ainsi que ses premiers cas d’usages. 📅 RDV le 10/10 à 14h ⚠️ Sur inscription 👉 ign.fr/agenda/confere… 🦾 L’#IGN mise sur l’#IA pour produire des données et

🚨The FLAIR dataset has been accepted at NeurIPS2023 in the Benchmark & Dataset track. It contains high-resolution aerial imagery with a ground sample distance of 20 cm and over 20 billion individually labeled pixels for precise land-cover classification. arxiv.org/abs/2310.13336

The FLAIR dataset 🛩️🛰️ is also available on Hugging Face : - over 20 billion annotated pixels of very high resolution aerial imagery at 0.2 m spatial resolution - along with Sentinel-2 satellite 1-year time series with 10 spectral band at 10m huggingface.co/datasets/IGNF/…