About

I am a Research Fellow at the Φ-Lab of the European Space Agency working on AI for Earth Observation. My research focuses on multimodal representation learning for Earth Observation, with an emphasis on foundation models. Before moving to ESA, I completed my PhD on multimodal self-supervised learning for Earth observation in Orion Lab, Athens supervised by Ioannis Papoutsis and Dimitrios Michail.

My work has been published at top ML conferences including ICCV, NeurIPS, and AAAI, and twice selected among UNESCO IRCAI’s Global Top 100 AI projects. You can find a full overview in my Resume and Publications.

News Feed

• June 29, 2026. I presented our work “From coordinates to context” at the UN World Food Programme workshop “Turning Models into Action: Advances in Modelling for Food Security and Humanitarian Response”
• Februrary 11, 2026, Our project Triton, was selected in the Global Top 100 AI projects list for 2025 of the International Research Center on Artificial Intelligence under the auspices of UNESCO.
• December 12, 2025, I succefully defended my PhD thesis on “Multimodal self-supervised learning for Earth observation”.
• October 19, 2025, Our paper Towards a Unified Copernicus Foundation Model for Earth Vision was selected as a Best Paper Award Finalist in ICCV 2025.
• September 10, 2025, Our paper Probabilistic Machine Learning for Noisy Labels in Earth Observation has been accepted for publication to Nature, Scientific Reports.
• July 25, 2025, Two papers accepted to ICCV 2025. On the Generalization of Representation Uncertainty in Earth Observation, joint work with Spyros Kondylatos and Towards a Unified Copernicus Foundation Model for Earth Vision (oral) led by Yi Wang.
• February 18, 2025, I’ll be a member of the scientific Committee for ESA-NASA International Workshop on AI Foundation Model for EO.
• December 9, 2025, Our paper FoMo: multi-modal, multi-scale and multi-task remote sensing foundation models for forest monitoring has been accepted to AAAI 2025, Artificial Intelligence for Social Impact Track.
• September 26, 2024, Our paper Kuro Siwo: 33 billion $m^2$ under the water. A global multi-temporal satellite dataset for rapid flood mapping has been accepted to NeurIPS 2024, Datasets and Benchmarks Track.
• March 27, 2024, Our paper FoMo-Bench: a multi-modal, multi-scale and multi-task Forest Monitoring Benchmark for remote sensing foundation models has been made available on arXiv.
• March 16, 2024 We have been selected as one of the Idea Track winners in EUSPA - EU Agency for the Space Programme Cassini Challenges with our idea manteo.
• November 18, 2023 Our paper Kuro Siwo: A global multi-temporal SAR dataset for rapid flood mapping has been made available on arXiV.
• October 3, 2023 We have received the “Best paper award” at the ICCV 2023 AI + HADR workshop for our work TeleViT: Teleconnection-driven Transformers Improve Subseasonal to Seasonal Wildfire Forecasting.
• August 21, 2023 Our work TeleViT: Teleconnection-driven Transformers Improve Subseasonal to Seasonal Wildfire Forecasting will be presented at the ICCV 2023 AI + HADR workshop.
• July 16, 2023: Orion Lab is holding a tutorial in IGARSS 2023 focusing on Deep Learning for monitoring and forecasting natural hazards with earth observation data.
• July 1, 2023: I’ll be acting as a Programme Committee for the 32nd International Conference on Artificial Neural Networks (ICANN).
• May 4, 2023: Our project Pluto has been selected for the International Research Center on Artificial Intelligence under the auspices of UNESCO Global Top 100 list 2022-23.
• April 1, 2023: I’ll be joining Mila - Quebec AI Institute as a research intern for the next 8 months.
• November 18, 2022: Our work Benchmarking and scaling of deep learning models for land cover image classification has been accepted for publication in ISPRS Journal of Photogrammetry and Remote Sensing.
• September 1, 2022: I join the AI4EO Future Lab of the Technical University of Munich as a Beyond Fellow.
• June 19, 2022: Our work Hephaestus: A large scale multitask dataset towards InSAR understanding has been published at the Computer Vision and Pattern Recognition (CVPR) workshop EarthVision.
• June 8, 2022: Our paper Learning From Synthetic InSAR With Vision Transformers: The Case of Volcanic Unrest Detection has been published in IEEE journal Transactions on Geoscience and Remote Sensing.
• August 19, 2021: Our paper Self-Supervised Contrastive Learning for Volcanic Unrest Detection has been published in IEEE journal Geoscience and Remote Sensing Letters.
• January 1, 2021: I join the Orion Lab of the National Observatory of Athens as a PhD candidate.