A Percentage Game: How AI, Image Recognition, and Motion Tracking Are Changing Tennis

Developments by IBM, Intel, SAP, the All England Lawn Tennis and Croquet Club (AELTC), and the Women’s Tennis Association have increased the accessibility of statistics and information surrounding tennis matches and are leading to changes in tennis’ ruleset, spectator experience, and high-level strategy.

 

Article Key

Image credit: IBM

Technology has given professional tennis a point advantage for almost two decades. As early as 2002, the Association of Tennis Players (ATP) and the Women’s Tennis Association (WTA) began relying on Hawk-Eye, a motion-tracking tool to determine whether a shot is in or out. The motion-tracking technology continued to gain popularity, being used in its first Grand Slam, the US Open, in 2006.

Today, almost all major tournaments have incorporated Hawk-Eye into the ruleset and flow of play, Roland Garros, the French Open, being the notable exception. Hawk-Eye makes use of six cameras located around the court to calculate the three-dimensional position of the tennis ball, accurate to the millimeter, at any given moment during a point. It’s used to review challenges to line calls so the chair umpire can determine a call’s accuracy.

In 2021, during the COVID-19 pandemic, Hawk-Eye was put on center court, as the Australian Open used it for all matches to protect line judges and limit possible exposure to the virus.

Image credit: Hawk-Eye

After almost two decades of state-of-the art replays, fans expect to watch multiple angles and tight footage of any shot in question. Hawk-Eye continues to add innovations to its motion-tracking technology to keep fans engaged and feeling as though they have front row seats to tennis’ hottest matchups.

AI Remasters Wimbledon Footage

After computer vision and motion-tracking technology’s successful introduction, the ATP and WTA added advanced analytics to give fans more of the stats they craved. The AELTC, a partner with IBM since 1990, added AI and image recognition beginning in the 2017 Wimbledon to automatically generate and pair match data with the pertinent match footage. 

IBM and AELTC also use audio sensors during Wimbledon to gauge audience feedback and shot sounds to create AI-generated highlight reels. The AELTC cannot televise all matches from Wimbledon due to the high volume of games in the early stage of the tournament, but through AI, image recognition, and audio sensors, it is able to provide broadcast quality highlights and insights for all matches and makes the highlights and analytics available on its website immediately following each match.

In 2020 when the world went topsy-turvy, the AELTC cancelled Wimbledon due to health concerns from COVID-19. For fans, as the first Wimbledon scratched since World War II, the moment was historic and depressing. AELTC and IBM made a smart pivot, though, and returned the COVID-19 backhand with a history-soaked, alternate Wimbledon experience.

Instead of live matches, the AELTC and IBM chose to produce and air “The Greatest Championships,” a remastered collection of the most memorable matches at Wimbledon. They turned to AI to update the footage, remaster sound and color, improve resolution, and refresh the matches with modern statistics.

 

Predictive Analytics Identifies Winning Patterns

In another return volley to the tennis world, SAP and the WTA took a different approach with analytics software, creating a resource to help players and coaches improve their game. SAP and the WTA encouraged players and coaches to use the new feature, Patterns of Play, to create and evolve match strategies.

Patterns of Play, a feature in SAP Tennis Analytics, provides insight into the probabilities and percentages behind each moment as a point develops through three forms of analysis:

  • Rally analysis informs the coach and player about each opponent’s specific hit location tendencies based on where the opponent is on the court.
  • Ball-toss analysis provides insight into the most common, successful ball-toss region.
  • Bounce-to-hit point analysis allows coaches and players to see serve-return success rates based on where the ball is returned to.

Intel® True View

Intel also has a play in immersive sports media experiences. Intel® True View is Intel Sports’ volumetric video platform for data capture, processing, and production. This new media format enables infinite storytelling from one capture. Using the volumetric capture technology, footage is recorded from dozens of 5K Ultra HD cameras to create a virtual environment in multi-perspective 3D, enabling fans to experience a moment on the pitch, court, or field from any angle.

The combination of analytics, AI, and motion-tracking video are changing the game for fans, players, and coaches. As many coaches will say, “Tennis is a percentage game.” The percentages are just more knowable now for everyone who loves the game.

 

Find Intel® IoT Market Ready Solutions

Search the Intel® IoT partner ecosystem and discover 130+ end-to-end IoT solutions.

Become a Member of the Intel® Partner Alliance

If you are delivering repeatable and scalable IoT offerings using Intel technology, the Intel® Partner Alliance program gives you more opportunities to build your business. Get started.