“If a bigger problem comes up in the future, it’s relatively easy for us to work with the installed base or the technical background we have at a facility and just add 10, 20, 30 or 40 different cameras,” he says. “We might want to focus them on certain parts of the field or use them for specific purposes.”
This type of scalability also brings to the table the concept of “digital twin” in sports. By capturing video footage and positioning data as a player moves on the field, that player can be virtually recreated – his movements, appearance and hand gestures all reproduced digitally in real time. This is something that has typically only been possible with the kinds of high-priced cameras and computer systems used in Hollywood and video game creation.
If digital twins can be created in sports, their uses go beyond refereeing. Announcers can use them in digital overlays that display statistics in real time or in virtual reality, so you can watch a game inside your VR headset.
Football is just the first playground for this technology. Almost any sport can benefit from the creation of digital twins, and Genius hopes to soon make inroads into basketball and American football.
But as intriguing as a digital twin of football sounds, can Dragon really solve the problems of offside detection? After all, the constant problems with previous VAR systems have not inspired confidence in motion capture technology among football’s key stakeholders or fans.
Genius says it has been testing Dragon for several years, both in the English Premier League and elsewhere, across multiple formats. The company employs several in-house analysts who project the tracking data into a video format and then analyse it frame by frame alongside the streamed video to spot any discrepancies. This allows the team to continually retrain its models until, theoretically, those errors are eliminated. Genius analysts consider this to be the fundamental level of testing, a baseline on which other layers are layered.
Dragon data has been compared to detection and VAR systems to validate its basic accuracy. It has also been manually validated – engineers spent long hours with various stakeholders in the sport (coaches, players, managers), running complex plays and confirming that the system’s results made sense. All clients considering using Dragon also have internal teams examining the system and validating its results.
“We’ve done it with groups like FIFA, where we’ve gone through extensive testing,” D’Auria says. “The Dragon system is validated by FIFA. They do tests where players wear a Vicon (motion capture) system, and we track them, and they compare data sets and look for errors. We’ve been through five or six iterations of this.”
It should be noted that both Genius and EPL representatives declined to provide specific test information or results to WIRED, stating that despite evaluating the iPhone system in parallel with VAR, comparisons to previous motion capture systems are complicated due to order-of-magnitude differences in the amount and quality of data being created. Interestingly, once again, both the EPL and Genius declined to give any indication as to how much more accurate their smartphone technology is compared to VAR.
Of course, the real assessment will be made by fans and players, who will have to see Dragon in action to believe it really makes a difference. The past few years of VAR nonsense have left an understandable bad taste in many people’s mouths when it comes to optical tracking.
But when the first semi-automated offside call comes this season in the UK, remember that this isn’t just the same old setup in a different wrapper. It’s the next generation of motion capture, one that stakeholders across all sports and the AI community will be watching closely. Fans won’t have much, if any, tolerance for any more issues with motion capture-based systems. Genius and the Premier League are confident they’re up to the challenge. We’ll see. Let the games begin.
