NFL hopefuls are adding AI video analysis to their arsenal

This site is reader-supported. When you click through links on our site, we may be compensated.

More than 130 football players have been training under the watchful eye of the athletic performance development company EXOS in Arizona, all in hopes of landing a first-round NFL draft pick. As it turns out, though, the eyes they’ve been working in front of aren’t exclusively human. Intel today said that EXOS’s latest batch of NFL hopefuls have been training in front of video cameras that — with the help of the company’s 3D athlete tracking system — should give players and staff a finer sense of their “body mechanics or trouble spots.”

“3DAT allows athletes to understand precisely what their body is doing while in motion, so they can precisely target where to make tweaks to get faster or better,” said Ashton Eaton, Intel product development engineer and two-time Olympic gold medalist.
The beauty of Intel’s 3DAT system is that athletes don’t need to strap on cumbersome sensors, or worry about precarious placement of gear during drills. Instead, run-of-the-mill video footage is shuttled off to servers packing Intel Xeon Scalable processors loaded with the company’s “Deep Learning Boost” AI acceleration capabilities. The system then tracks 22 distinct points on an athlete’s body, and analyzes their form for velocity, body angles, and acceleration points. Finally, those results are relayed back to training staff in the form of chart-laden reports, all designed to help players better understand their running technique and ways it could be improved.

“This data enables us to make adjustments in the weight room to help unlock more potential on the field,” said Craig Friedman, SVP president of EXOS’ Performance Innovation Team, in a press release.

This tie-up with EXOS might be the most notable use of Intel’s athlete tracking tech to date, but it wasn’t always going to be that way. Viewers around the world were set to get a first-hand look during the 2020 Tokyo Olympics, where it would’ve overlaid algorithmically generation visuals of athletes’ forms over replays of events like the 100-meter dash. The worldwide COVID pandemic shuttered plans to stage the event last year, but such a high-profile tech demo may still be used when the Olympics kick off later this year. 

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.