Call for Abstracts
The development of Artificial Intelligence (AI) techniques for Earth Observation (EO) is leading to significant advances in environmental monitoring, climate science, and sustainable resource management. AI techniques are enhancing the extraction and analysis of information from satellite, aerial, and in-situ data, enabling more accurate, scalable, and timely insights into Earth system dynamics. This call invites original and new research contributions that explore methodological innovations, theoretical advancements, and applied studies at the intersection of AI and EO. We accept a variety of applications, including but not exclusive to environmental monitoring, climate change, biodiversity, urban planning, agriculture and disaster management.
Relevant topics include, but are not limited to:
- Advanced learning strategies, including un/semi/self-supervised learning, few-shot learning, and continual learning
- Novel geospatial foundation models and their evaluation
- Multi-modal data analysis, including textual data and large language models
- Physics-based machine and deep learning
- Data fusion techniques
- Analysis of complex data, including SAR, hyperspectral, or 3D point clouds data
- Super-resolution of satellite images
- Multi-temporal data analysis and change detection
- Semantic understanding, including classification, semantic segmentation, and object detection
- Big data and HPC
- Uncertainty quantification and explainable AI
- Applications in climate change, biodiversity, environmental monitoring, and disaster management or other EO-related topics
- New original benchmarks
- Embedded and frugal AI for Earth observation
- Cybersecurity challenges, including integrity and confidentiality
Each abstract should include the following information to undergo the rigorous single-blind peer-review process by the Scientific Committee:
- Title
- List of authors and their affiliations
- List of keywords (5 maximum)
- Summary of the abstract (150 words) to be displayed in the online programme
- Abstract (500 words or 1 page, PDF format), including presentation of the context, research objectives, data, main (expected) results, conclusions and perspectives
The deadline for abstract submission is 28 May 2025. The acceptance of the abstract for oral or poster presentation will be based on the originality, significance, technical soundness, and clarity of presentation. There will be no formal proceeding.
Please note that we also accept abstract submission of papers presented in main Machine Learning and Computer Vision conferences during the last year (January 2024 to now), notably CVPR, ECCV, ICML, ICLR, NeurIPS, AAAI, or ECML-PKDD.
To submit a paper, please use this CMT submission link.
The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.
