Foundation and Explainable Artificial Intelligence for Visual Analysis and Semantic Querying of Marine Environments

The research activity focuses on the development of advanced Artificial Intelligence methods for the automatic analysis of underwater images and videos aimed at supporting marine biodiversity monitoring. Foundation models for species recognition, habitat analysis, and environmental anomaly detection will be developed by integrating self-supervised learning, domain adaptation, and Explainable AI techniques. The research will also address semantic querying through Large Language Models and dedicated preprocessing pipelines for underwater visual data.