CHICAGO – March 4, 2019 –– Today, STATS, the worldwide leader in sports data and intelligence, announced the official launch of AutoSTATS, the first artificial intelligence (AI) and computer vision technology to deliver comprehensive player-tracking data through any broadcast.
AutoSTATS harnesses the latest capabilities in AI and machine learning to automate tracking with X/Y coordinates, differentiating players and the ball and coordinating player movements around the court or playing field. The technology has the ability to track deep player and team performance data directly from video, opening the door to track historical games when optical tracking solutions weren’t available.
“In 2009, STATS became the first company to install optical player tracking cameras in the rafters of NBA, capturing the deepest player-tracking and team data at that time,” said Dr. Helen Sun, Chief Technology Officer at STATS. “Since we first installed our SportVU technology, tracking data has become an integral part of game preparations for coaches, analysts, media and fans in almost every professional sport. However, with player tracking only available in stadiums with optical tracking cameras, there is an enormous gap in its availability.”
AutoSTATS has already begun collecting tracking data for select college basketball games during the 2018-19 season, where optical tracking solutions are limited. As a result, the Orlando Magic recently signed an exclusive deal with STATS, making the Magic the only NBA team with access to AutoSTATS data, which will be used for college scouting.
STATS has also begun adding a new layer to AutoSTATS using OpenPose, a product manufactured under license from Carnegie Mellon University. OpenPose unlocks new layers of body-pose information, providing a deeper quality of player tracking data like body position, shot form, torque, and other aspects of the game.
“STATS has been lucky to have the deepest trove of sports data, as well as the most player tracking data in the world—dating back to the first optically tracked soccer game in 1999,” said Dr. Patrick Lucey, VP of AI at STATS. “Using our years of AI and machine learning research, we have been able to plug that data into AutoSTATS, teaching the computer new aspects of the game faster. This new technology has the ability to unlock more data and insights than any sports company has ever been able to collect.”
STATS’ award-winning AI team has published 16 papers and a host of patents since 2016. The team and research has been recognized at the world’s most prestigious AI conferences.
“AutoSTATS is a completely new technology, that wouldn’t be available without developing the robust AI, computer vision and machine learning technologies we have in place today,” said Dr. Sun. “While STATS started in 1981 as a small sports data and analytics company, we have become an AI-centric Software as a Business company, improving fan engagement, sports betting, team performance and computer vision based data generation. Our vision and product strategy is centered on a culture of fearless innovation that redefines how sports are consumed, understood and played. I couldn’t be more proud of how AutoSTATS epitomizes that drive.”
For more information about AutoSTATS, visit: https://www.stats.com/auto-stats/.