IBM's Watson will speed-edit Wimbledon videos by recognising player emotions

In this year's tie-up with The All England Lawn Tennis Club, IBM's AI platform, Watson, will be speeding the video editing of Wimbledon footage by spotting "player emotions".

The Watson-powered Wimbledon dashboard
The Watson-powered Wimbledon dashboard

With an average of three matches per court, per day, video from Wimbledon's matches can quickly add up to hundreds of hours of footage which could take hours to pull together into highlight packages.

This year, Watson has been taught to recognise video highlight moments by analysing crowd noise, players’ movements and match data. This allows the Wimbledon editorial team to scale and accelerate the video production process for highlight packages and expand the number of potential matches that are turned into timely highlight videos for fans to watch and share.

This has reduced the time needed to process videos from 15 minutes last year to five minutes this year, not including the time it takes for manual content review.

Accompanying this is an AI highlights dashboard that will populate in near real-time every shot of a match and its excitement level. The AELTC digital team will be able to view and find the most exciting shot of the day or the match and leverage this content across all their digital channels, including social.

Updated Slamtracker and chatbot

IBM has also updated its tennis analysis function, SlamTracker. It is now capable of providing varying levels of commentary and analysis to suit different fans - from high-level overviews to point-by-point commentary.  Real-time data will be integrated from multiple sources including courtside statisticians, chair umpires, radar guns, ball position, player location and even Twitter for social sentiment.

IBM’s SlamTracker works by analysing the players’ cognitive "keys" to understand which tactics to look for in a head to head match. The aim is to reveal the hidden patterns in player and match dynamics. The solution also uses Watson APIs to refine and update the player style based on match data. SlamTracker will also now have an added feature referred to as "momentum" which will be a visual depiction of a match with an indication of which player has the momentum and how that may have shifted over the course of the match.

Wimbledon's AI-assistant chatbot, Fred, has also received an upgrade and will be available on Facebook Messenger. Fans can ask Fred about scores, news and players while driving engagement back to the official platforms.