AI Case Study
US Open speeds up tennis match highlight availability on its app by 2-10 hours through automating video selection process
The US Open has implemented IBM Watson's AI to automatically parse footage of tennis matches to choose highlights. It does so by ranking moments according to audio data, such as crowds cheering and visual data and then selecting the highest ranking highlights to feature on its app. This feature selection automation allowed highlights to be shown up to 10 hours earlier than under manual selection.
Consumer Goods And Services
Entertainment And Sports
According to Inc., IBM Watson "will help the USTA sort through thousands of highlights each day, selecting the best ones using its own scoring system. Watson's algorithm will decide which clips are worthy, and those videos will be pushed directly to the phones of those who downloaded the US Open app and opted in. Watson's new system, called Cognitive Highlights, measures the volume of the crowd (which is known to be raucous), the commentators' analysis, and the players' reactions. It assigns each of those a score from 0 to 1, and then uses those inputs to determine an Overall Excitement score. A point that results in a player yelling and fist pumping, then, will generate a higher score than one in which she pats her racket against her leg. The system can also parse the analysts' language, so a "solid" shot won't score as well as an "awesome" one."
Creative And Brand
From Inc.: "The US Open pits 256 players against one another, competing for more than $50 million in prize money in this year's Grand Slam event. A quarter-million points will be scored during the two-week tournament in Queens, in New York City, and at any given time, 17 matches can be in progress. To make sense of all that, and serve up better, quicker digital access, the U.S. Tennis Association is using IBM Watson."
Per IBM, due to the video selection automation "highlights and key moments were available to fans two to 10 hours more quickly than during previous years".
"The system uses deep learning models and “self-supervised” active learning techniques to recognize which of these points are significant and understand what makes a good highlight. Using Watson APIs, it understands the importance of certain scores—like a point that clinches a set—and uses visual and audio cues to create ratings for each point. The system also uses facial recognition to read the emotional reactions of the players. For sound classification, the team worked with MIT to develop a deep neural network called “SoundNet” for environmental sound analysis like crowd noise." (IBM developerWorks)
Audio and visual data to rate the "over 25,000 points from over 300 hours of coverage of men and women’s singles matches". (IBM developerWorks)