Solène Debuysère

Information

Bio

I am Solène Debuysère, a final-year PhD student at ONERA – The French Aerospace Lab, affiliated with the EOBE Doctoral School at Université Paris-Saclay. My research focuses on multimodal generative models for Very High-Resolution Synthetic Aperture Radar (VHR SAR) imagery, with a particular interest in fundamental questions related to statistical modeling approaches for SAR data.

Current Role

My work explores the adaptation of latent diffusion and foundation models to the SAR domain. An initial research direction focuses on amplitude mono-polarization data acquired in X-band using the airborne ONERA SETHI sensor. The objective is to leverage language semantics to describe and generate coherent radar scenes while integrating physical constraints to ensure geometric and radiometric consistency in the generated imagery. A second line of research extends this approach to full polarimetric learning, aiming to capture richer scattering information and underlying physical structure.

Areas of Expertise

My expertise lies at the intersection of Artificial Intelligence and Earth Observation, including generative modeling, multimodal learning, SAR image understanding, and physically informed deep learning models. These specializations allow me to investigate how data-driven models can remain consistent with radar imaging physics while enabling controllable scene generation and improved data representation.

Vision for the Future

My long-term goal is to support the development of models that provide more reliable, interpretable, controllable, and accessible analysis of complex SAR imagery for scientific and operational applications.

Project as CIN Researcher

Coming Soon