Search

Image Details

View Article Bookmark & Share

Choose export citation format:

Export Citation
Large resolution
Hi-Resolution Export to PDF

Astronomical Image Time Series Segmentation for Faint, Fast Object Detection

  • Authors: Dan Caselden, J. Davy Kirkpatrick, Lindsey Lack, Andrew White, Federico Marocco, Aaron M. Meisner, Guillaume Colin, Bruce Baller, Kareem Ammar, Thomas P. Bickle, Hunter Brooks, S. L. Casewell, Peter R. M. Eisenhardt, Charles A. Elachi, Jacqueline K. Faherty, John W. Fowler, Daniella C. Bardalez Gagliuffi, Jonathan Gagné, Christopher R. Gelino, Jake Grigorian, Les Hamlet, Hiro Higashimura, 村滉 東, Justin Hong, Issam Ibnouhsein, Tarun Kota, Marc J. Kuchner, Ken A. Marsh, Matteo Paz, Yadukrishna Raghu, Adam C. Schneider, Sajesh Singh, Asa Trek, Kieran Wall, Andrew Washburn, Edward L. Wright, David Zurek

Dan Caselden et al 2026 The Astronomical Journal 171 .

  • Provider: AAS Journals

Caption: Figure 3.

Validation loss during SMDET training (Section 3). This loss was calculated on a held-out validation dataset to estimate its performance on unseen data. The x-axis represents millions of processed minibatches. The series depict object mask (yobj) loss, segmentation mask (yseg) loss, and their average that we used for model updates.

Other Images in This Article
Image in this article
Image in this article
Image in this article
Image in this article
Image in this article
Image in this article
Image in this article
Copyright and Terms & Conditions

Additional terms of reuse