NFL and AWS Challenge AI to Find Ways to Automate Player Identification Using NFL Play Sequences
$ 100,000 in total price available; open until November 2, 2021
The National Football League (NFL) and Amazon Web Services (AWS) have launched a new artificial intelligence Kaggle challenge to create ways for computers to automatically identify players using NFL play footage. New computer vision models created through the challenge will accelerate the NFL’s work with AWS to better understand and aim to reduce injuries in the NFL. Cash prizes totaling $ 100,000 will be awarded to data scientists with the winning models. The competition will be open until November 2, 2021.
This challenge is the next step in the NFL’s work with AWS to develop the Digital Athlete, a virtual representation of an NFL player that can be used to better predict and hopefully prevent player injuries. Last season, the NFL hosted its first computer vision competition, providing computer scientists with NFL game data and challenging them to create ways to detect helmet impacts in the field. Almost 7,800 submissions were received from data scientists around the world. These solutions are now being used by the NFL and AWS in ongoing work on the digital athlete.
This challenge builds on last season’s competition by scoring teams based on how accurately they identify players involved in a helmet impact. The NFL will provide data scientists with the same set of gaming data to build on current models. This competition will advance an already strong AWS-based machine learning pipeline that strives to be among the most advanced and sophisticated injury surveillance and mitigation programs in professional sports.
“This competition is fundamental in helping to identify each player’s risk of injury-causing events, particularly with regards to head health,” said Jeff Miller, NFL executive vice president overseeing programs. league health and safety. “We are excited to recruit some of the most talented data scientists from around the world to help develop these solutions and significantly advance player security. “
“Using AWS’s wide range of technologies, including machine learning (ML) and artificial intelligence (AI), the NFL and AWS will generate more new information than ever before about player injuries, the rules of the game. play, equipment, rehabilitation and recovery, ”Priya said. Ponnapalli, Senior Director, Applied Sciences, Amazon ML Solutions Lab. “This competition will continue that work and the data and information gathered through this project has the potential to shape the future of football.”
The digital athlete will ultimately help the NFL and its clubs leverage sports science and injury biomechanics to develop individualized training and recovery regimes, perform real-time injury risk analysis during games , and identify and evaluate other player safety initiatives, including around equipment, rule changes and coaching techniques.
To enter the competition, go to https://www.kaggle.com/c/nfl-health-and-safety-helmet-assignment/ for more information on rules and eligibility.
About NFL player health and safety initiatives
The NFL is committed to advancing the diagnosis, prevention and treatment of sports-related injuries. The league’s approach to player health and safety is rooted in the collection and analysis of data to help guide its ongoing efforts to understand and reduce injury, and protect players. Data and science underpin the NFL’s efforts to further improve medical protocols, improve the way our game is taught and played, and spur the development of new and improved protective gear, including helmets. For more information on the NFL’s health and safety efforts, please visit www.NFL.com/PlayerHealthAndSafety.
Sign up for the free insideBIGDATA newsletter.
Join us on Twitter: @ InsideBigData1 – https://twitter.com/InsideBigData1