Vincent Miele

The paper analyzes changes in bird interaction networks across North America over six decades, revealing shifts driven by species abundance and human activity.

This paper improves camera trap classification models by incorporating self-attention to leverage information from multiple animal detections, enhancing species classification accuracy in both mono- and multi-species scenarios.

This study evaluates Earth Observation Foundation models for predicting soil biodiversity in the French Alps, finding that high-quality in-situ data outperform remote sensing embeddings, though they offer useful spatial context.

This study uses vision-language foundation models to classify animal behaviors in camera trap images without prior training, achieving over 91% accuracy in three ungulate species.