top of page

OK COMPUTER VISION

Updated: Jun 5, 2023


Title page for OK Computer Vision blog post

I’ve spent my last four years at Sportradar working with a form of artificial intelligence called Computer Vision which teaches computers to see and understand visual inputs such as digital images or videos. Or, to put it simply, it mimics human vision capabilities.

And while we, as individuals, are very good at understanding what is happening in front of us at sporting events, we have limited capacity in the amount of data we are able to manually record and upload. By comparison, Computer Vision is capable of capturing a 100-fold increase in the level of statistics and data it collects, in milliseconds.

It’s this super-fast, super-deep and contextualised data that’s going to transform the industry and change how athletes, fans and bettors experience and engage with sport, and Computer Vision is the technology to enable this at scale.

That transformation is already well underway and earlier this month Sportradar demonstrated its Computer Vision capabilities for the first time with a table tennis showcase at ICE London, the betting industry trade show.

The event was the perfect opportunity for us to show the industry that we have a fully functioning Computer Vision solution. We’ve trained the technology to understand the rules of table tennis, detect all game events in a fraction of a second, and can deploy it at scale across a table tennis portfolio that includes more than 190,000 matches this year.



What’s more significant is that the showcase provided a glimpse of how we’ll all experience sport in the future.

For the athletes competing in events, Computer Vision provides them with the data and insights they need to improve performance. The depth of data generated by Computer Vision is going to shape coaching methods, focus training sessions and inform match tactics. Additionally, real-time performance reports that include details on every serve, stroke and rally can influence how an athlete plays a match.


Computer Vision in action at ICE conference

We’ll see it improve the integrity of sports and alert to doping amongst athletes. Every athlete has a unique, signature style of play and certain characteristics, which can be tracked over time. That signature will naturally evolve over their career, but unnatural changes can also be identified. If there’s suddenly a marked change in performance on these mechanical characteristics, it provides valuable insight into suspicious activity.

For the fans, Computer Vision is going to immerse them more deeply into the sports they love. The technology turns every video frame into a digital asset, which we’ll use to transform what they see on their screens. As a result, fans can expect to see dynamic content such as player performance insights, live betting data incorporated into a live stream, and even things such as 3D replays from alternative camera angles, which I’m confident will all have a positive effect on viewing time.

Using this technology to transform streaming is going to help rights holders to ‘premiumise’ their content and open new opportunities for advertisers and sponsors to engage fans. The advanced digital nature of the technology unlocks significant activation opportunities for brands with animations and dynamic, contextually relevant, even geo-tagged LED advertising boards updated in real-time amongst some of the options available to those forward-thinking brands.



Computer Vision in action at ICE conference

There’s tremendous opportunity to unlock here and I believe where we are now is just the tip of the iceberg. Consider the application of this technology within the metaverse or Virtual Reality. In the future we’ll be able to provide fans with an avatar that has real parameters. We can make fans part of a professional tennis match, for example, and immerse them in a real virtualised match or even allow them to participate in a game.

Technology allows us to collect deep data, right down to player skeletons, which takes the idea of an immersive fan experience to a whole new level because we can simulate matches and their outcomes, matching players that otherwise wouldn’t be playing against each other and more.

I truly believe that Computer Vision will transform the sports ecosystem and I’m excited to see the opportunities it creates for all those within it.


Luka Pataky is Chief Product Officer - Computer Vision for Sportradar.

278 views0 comments

Comments

Rated 0 out of 5 stars.
No ratings yet

Add a rating
bottom of page