Celeste: Variational inference for a generative model of astronomical images
This work addresses the challenge of accurate astronomical image analysis for researchers, though it appears incremental as it builds on existing generative and variational inference methods.
The authors tackled the problem of modeling optical telescope image sets with a fully generative approach, achieving performance that exceeds the current state-of-the-art method for locating celestial bodies and measuring their colors in a major sky survey.
We present a new, fully generative model of optical telescope image sets, along with a variational procedure for inference. Each pixel intensity is treated as a Poisson random variable, with a rate parameter dependent on latent properties of stars and galaxies. Key latent properties are themselves random, with scientific prior distributions constructed from large ancillary data sets. We check our approach on synthetic images. We also run it on images from a major sky survey, where it exceeds the performance of the current state-of-the-art method for locating celestial bodies and measuring their colors.