Applying Vision Transformers on Spectral Analysis of Astronomical Objects

TL;DR

This paper demonstrates that pre-trained Vision Transformers can effectively analyze astronomical spectral data by converting spectra into images, achieving high accuracy in classification and redshift estimation.

Contribution

It introduces a novel approach of applying pre-trained Vision Transformers to spectral analysis by transforming spectra into images, showing superior performance over traditional methods.

Findings

Achieved higher classification accuracy than SVMs and Random Forests.

Attained R^2 values comparable to specialized spectrum encoders.

Validated effectiveness on large-scale real spectroscopic data.

Abstract

We apply pre-trained Vision Transformers (ViTs), originally developed for image recognition, to the analysis of astronomical spectral data. By converting traditional one-dimensional spectra into two-dimensional image representations, we enable ViTs to capture both local and global spectral features through spatial self-attention. We fine-tune a ViT pretrained on ImageNet using millions of spectra from the SDSS and LAMOST surveys, represented as spectral plots. Our model is evaluated on key tasks including stellar object classification and redshift ($z$) estimation, where it demonstrates strong performance and scalability. We achieve classification accuracy higher than Support Vector Machines and Random Forests, and attain $R^2$ values comparable to AstroCLIP's spectrum encoder, even when generalizing across diverse object types. These results demonstrate the effectiveness of using pretrained vision models for spectroscopic data analysis. To our knowledge, this is the first application of ViTs to large-scale, which also leverages real spectroscopic data and does not rely on synthetic inputs.