Deep Learning for Automatic Basketball Video Production with Julian Quiroga - #389
Today we return to our coverage of the 2020 CVPR conference with a conversation with Julian Quiroga, a Computer Vision Team Lead at Genius Sports.
Julian presented his recent paper “As Seen on TV: Automatic Basketball Video Production using Gaussian-based Actionness and Game States Recognition” at the CVSports workshop. We jump right into the paper, discussing details like camera setups and angles, detection and localization of the figures on the court (players, refs, and of course, the ball), and the role that deep learning plays in the process. We also break down how this work applies to different sports, and the ways that Julian is looking to improve on this work for better accuracy.
Thanks again to our friends at Qualcomm for their support of the podcast and sponsorship of this series!