Comment spécialiser DINOv2 pour l'astronomie?

Jun 1, 2025·
Matthieu Le Lain
Matthieu Le Lain
· 1 min read
Abstract
This study evaluates the performance of existing visual foundation models, based on ViT, SwinV2, BEiT or DINOv2, for astronomical applications, particularly galaxy morphological classification using the Galaxy10 DECaLS dataset. We explore different fine-tuning strategies to specialize DINOv2 for astronomy.
Date
Jun 1, 2025
Location

France

events

Talk presented at GRETSI 2025. This work evaluates the performance of existing visual foundation models (ViT, SwinV2, BEiT, DINOv2) for astronomical applications, focusing on fine-tuning strategies to specialize DINOv2 for galaxy morphological classification.

Foundation Models DINOv2 Galaxy Classification Talk
Matthieu Le Lain
Authors
Matthieu Le Lain
AI for astronomy & astrophysics
PhD student at IRISA, Université Bretagne Sud (expected 2026), and lecturer in computer science, working on foundation models for astronomy and astrophysics, with a broader interest in deep learning applied to scientific data. Also contributing to the UniverseTBD collaboration on vision–language models for astronomy.