FAST-EO

?? FAST-EO consortium just met at ESA/ESRIN (Frascati, Italy) to discuss project progress ?? [Press Release and Project Website now online!] ? FAST-EO, funded by the European Space Agency - ESA Φ-lab (Nicolas Longépé, Nikolaos Dionelis), is dedicated to advancing AI for Earth observation, with a focus on the integration and efficiency of multi-modal geospatial foundation models. ? The consortium, composed of the German Aerospace Center (DLR) (R?dvan Salih, Corneliu Octavian Dumitru, Antony Z.), Forschungszentrum Jülich (Gabriele Cavallaro, Rocco Sedona, Kennedy Adriko, Stefano Maurogiovanni, Stefan Kesselheim, Alexandre Strube), IBM Research (Benedikt Blumenstiel, Thomas Brunschwiler, Johannes Jakubik, Katia Moskvitch, Romeo Kienzler), and KP Labs (Jakub Nalepa, Agata Wijata, ?ukasz Tulczyjew), gathered at ESA/ESRIN to review the project's activities. ? ?? Read the press release https://lnkd.in/e6rZz6WD ?? Discover objectives and use cases on our website https://www.fast-eo.eu/ ? #foundationmodels #ai #eo #remotesensing #europeanspaceagency

?? The FAST-EO European Space Agency - ESA Ф-lab project was pleased to participate in the "Foundation Models for Earth Observation" workshop organized in collaboration with the Helmholtz AI and the International Future Lab AI4EO The event included a panel discussion on the role of AI Foundation Models in Earth Observation. FAST-EO conducted a hands-on session showing how to fine-tune the open-source geospatial foundation model, Prithvi, on own data for specific tasks, using the JURECA supercomputer operated by the Jülich Supercomputing Centre (JSC) Forschungszentrum Jülich Thanks to the organizers for inviting us and to Thomas Brunschwiler, Benedikt Blumenstiel (IBM Research), and Stefano Maurogiovanni (Forschungszentrum Jülich) for representing FAST-EO Event: https://lnkd.in/eBq2BDA2 Learn more about FAST-EO: https://lnkd.in/eq3khGnZ #ai #ai4eo #foundationmodels #panel #project #remotesensing

Thrilled to have been part of the #EcoHackathon kickoff at #ETH! Can't wait to see the innovative solutions the talented participants develop using our Earth Observation Foundation Model #Prithvi to assess biodiversity in the Amazon rainforest. Looking forward to the celebration on June 10th and hopeful that #Prithvi will be a contender in the #XPRIZE finals. Big thanks to David Dao for including us in the hackathon and to Carlos Gomes for the mentoring on how to fine-tune #Prithvi with our latest library #TerraTorch! EO-FM Prithvi https://lnkd.in/eMtf-zuv TerraTorch https://lnkd.in/eEicn9Sa

Exciting news for the future of AI Foundation Models for Earth observation!?? ?? Happy to announce that on February 5th we kicked off the European Space Agency - ESA FAST-EO project "Fostering Advancements in Foundation Models via Unsupervised and Self-supervised Learning for Downstream Tasks in Earth Observation" with Bertrand Le Saux, Nicolas Longépé, Giuseppe Borghi, et al. The project’s aim is to enhance the accessibility and democratization of FMs within the EO community with the following actions: tailoring to data domains, enhanced multimodality, overcoming computational barriers, enable affordable fine-tuning, provide operational and accessible implementations. Under the leadership of German Aerospace Center (DLR) (R?dvan Salih, Corneliu Octavian Dumitru, et al.), the consortium, which includes Forschungszentrum Jülich (Gabriele Cavallaro, Stefan Kesselheim, Rocco Sedona, et al.), KP Labs (Jakub Nalepa, ?ukasz Tulczyjew, Agata Wijata, et al.) and IBM Research (Thomas Brunschwiler, Johannes Jakubik, Romeo Kienzler, et al.) gathered to discuss the project's objectives and set the stage for our future efforts. Stay tuned for updates! ?? #ai #ai4eo #foundationmodels #kickoff #project #remotesensing

Exciting news for the future of AI Foundation Models for Earth observation!?? ?? Happy to announce that on February 5th we kicked off the European Space Agency - ESA FAST-EO project "Fostering Advancements in Foundation Models via Unsupervised and Self-supervised Learning for Downstream Tasks in Earth Observation" with Bertrand Le Saux, Nicolas Longépé, Giuseppe Borghi, et al. The project’s aim is to enhance the accessibility and democratization of FMs within the EO community with the following actions: tailoring to data domains, enhanced multimodality, overcoming computational barriers, enable affordable fine-tuning, provide operational and accessible implementations. Under the leadership of German Aerospace Center (DLR) (R?dvan Salih, Corneliu Octavian Dumitru, et al.), the consortium, which includes Forschungszentrum Jülich (Gabriele Cavallaro, Stefan Kesselheim, Rocco Sedona, et al.), KP Labs (Jakub Nalepa, ?ukasz Tulczyjew, Agata Wijata, et al.) and IBM Research (Thomas Brunschwiler, Johannes Jakubik, Romeo Kienzler, et al.) gathered to discuss the project's objectives and set the stage for our future efforts. Stay tuned for updates! ?? #ai #ai4eo #foundationmodels #kickoff #project #remotesensing