New artificial intelligence technology developed by engineers at Brigham Young University could significantly reduce the time and cost of film study for Super Bowl-bound teams (and all NFL and college football teams) while improving game strategy by leveraging the power of big data.
BYU Professor D.J. Lee, master’s student Jacob Newman and Ph.D. students Andrew Sumsion and Shad Torrie use artificial intelligence to automate the time-consuming process of manually analyzing and annotating game footage. Using deep learning and computer vision, the researchers created an algorithm that can consistently locate and tag players from game film and determine the formation of an offensive team, a process that can take the time of multiple video assistants.
Lee, a professor of electrical and computer engineering says “We had a conversation about it and realized that we could probably teach an algorithm how to do this, so we set up a meeting with BYU Football to learn about their process and immediately knew, yes, we can do this much faster”.
While the research is still early days, the team has already achieved better than 90% accuracy in detecting and tagging players with their algorithm, along with 85% accuracy in determining formations. He believes the technology could eventually eliminate the need for the inefficient and tedious manual annotation and analysis of recorded video used by NFL and college teams.
Additional angles to properly train algorithm
Lee and Newman got their first look at actual game footage provided by the BYU football team. When they started analyzing it, they realized they needed some additional angles to properly train their algorithm. So they bought a copy of Madden 2020, which shows the field from above and behind the offense, and manually tagged 1,000 images and videos from the game.
They used these images to train a deep learning algorithm to find players, which would then feed into the Residual Network framework to determine what position the players were playing. Finally, their neural network uses location and position information to determine what formation (of over 25 formations) the offense is using – anything from a Pistol Bunch TE to an I Form H Slot Open.
Lee said the algorithm can accurately identify formations 99.5% of the time when player placement and marking information is correct. The I formation, where four players are lined up in front of each other – center, point guard, fullback and defender – has proven to be one of the most difficult formations to identify.
Lee and Newman said the AI system could have applications in other sports as well. In baseball, for example, it could locate players’ positions on the field and identify common patterns to help teams improve how they defend against certain batters. Or it could be used to scout football players to help determine more efficient and effective formations.
“Once you have that data, you can do a lot more with it; you can take it to the next level,” Lee said. “Big data can help us understand the strategies of this team or the tendencies of this coach. It could help you figure out if they’re likely to go for it on 4th Down and 2 or if they’re going to punt. The idea of using AI for sports is really cool, and if we can give them even a 1% advantage, it will be worth it.”
Read Now :<strong>Mars Wrigley India expects to double its stores every 5 years till 2042</strong>
