Galaxy Morphological Classification with Deep Learning
Nov 15, 2024·
·
1 min read
Matthieu Le Lain
Galaxy morphological classification — spiral, elliptical, irregular? — is one of the most natural applications of computer vision in astronomy. Datasets like Galaxy Zoo and Galaxy10 DECaLS provide crowd-sourced morphological labels for thousands of galaxies and serve as standard benchmarks.
In my work with Sébastien Lefèvre, we compared several pre-trained foundation models (ViT, SwinV2, BEiT, DINOv2) on Galaxy10 DECaLS to see how well they transfer to astronomical images. The short answer: surprisingly well, but the fine-tuning strategy matters a lot. These results were presented at EAS 2024 and GRETSI 2025, with a detailed analysis in the companion paper.
Authors
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.