AI system analyses live video to create data for graphics and gambling products
Sportradar has launched Computer Vision, a live tracking system for sports that utilises AI software.
Computer Vision can analyse a live video feed at 120 frames per second, and collect 100 times the data that is usually collected in that time by training computers to see and understand visual inputs from the video and images. It can create the data using just one camera for table tennis, or scale this up for larger playing areas, and be set up by a small team in just two hours. Data can include scoring, skeletal tracking, ball tracking, and more.
This data can then be fed into other Sportradar products and create live graphics, 3D replays, virtual advertising, gambling data, and more.
The company has launched Computer Vision at the ICE conference this week, and is currently rolling it out across its table tennis partners, with tennis to be the next sport that Sportradar targets. The technology is adaptable to any sport, with US sports expected to be a growth area for its use.
Luka Pataky, chief product officer for Computer Vision at Sportradar, said: “The table tennis showcase is the first time that Sportradar has demonstrated its Computer Vision technology in this way. We believe that this AI-driven tech has the potential to transform how fans engage and experience sport now and in the future. From a streaming perspective, the technology is capable of transforming what fans see on their screens, incorporating match data, performance insights and betting markets into a live stream, providing viewers with unique engagement opportunities.”