We present Tails, an open-source deep-learning framework for the identification and localization of comets in the image data of the Zwicky Transient Facility (ZTF), a robotic optical time-domain survey currently in operation at the Palomar Observatory in California, USA. Tails employs a custom EfficientDet-based architecture and is capable of finding comets in single images in near real time, rather than requiring multiple epochs as with traditional methods. The system achieves state-of-the-art performance with 99% recall, 0.01% false positive rate, and 1-2 pixel root mean square error in the predicted position. We report the initial results of the Tails efficiency evaluation in a production setting on the data of the ZTF Twilight survey, including the first AI-assisted discovery of a comet (C/2020 T2) and the recovery of a comet (P/2016 J3 = P/2021 A3).
@article{arxiv.2102.13352,
title = {Tails: Chasing Comets with the Zwicky Transient Facility and Deep Learning},
author = {Dmitry A. Duev and Bryce T. Bolin and Matthew J. Graham and Michael S. P. Kelley and Ashish Mahabal and Eric C. Bellm and Michael W. Coughlin and Richard Dekany and George Helou and Shrinivas R. Kulkarni and Frank J. Masci and Thomas A. Prince and Reed Riddle and Maayane T. Soumagnac and Stéfan J. van der Walt},
journal= {arXiv preprint arXiv:2102.13352},
year = {2021}
}