FIFA’s VAR System and using AI in sporting tournaments

The Virtual Assistant Referee, also known as VAR, has certainly stolen the limelight at the 2018 World Cup so far. Numerous people would agree it has had a mostly positive effect, leading to fewer red cards per game after each country’s first fixture than at any World Cup for 32 years. Video analytics could soon find a more permanent place in football, as well as many other areas. But with the speed of play being questioned by some due to the use of VAR, how could this be drastically improved?  

The concept behind VAR

The VAR actually consists of a team of people who watch video replays of relevant incidents and work together to review certain decisions made by the main referee. That team comprises four referees who are glued to video screens during each match: one video assistant referee (the main VAR) and three assistant video assistant referees (AVAR1, AVAR2 and AVAR3). The team are located in a video operation room (VOR), which has a row of monitors showing different camera angles. Each of the 12 stadiums’ video feeds are sent to the VOR (and each stadium has 33 cameras) through fibre optics.

“The use of artificial intelligence (AI) in sporting events is already a reality, and it has so much more potential”

The main VAR watches one video feed, which has a monitor underneath comprising split-screen that shows four feeds. AVAR1 takes the main feed, AVAR2 watches two screens which stream two dedicated cameras that aren’t used anywhere else, and is in charge of offside challenges; and AVAR3 is watching the game through the same television feed we watch in our homes.  

Is it having a positive effect?

The process for reviewing a decision can work in two ways; either the referee on pitch can request a review after making a decision or the VAR team can recommend one. If the VAR judges that there is the potential for a clear error to have been made then they can notify the referee. Everyone has an opinion on VAR, but overall, it appears most people believe it is having a positive impact on the game. However, despite the accuracy of VAR (99.3%), many are complaining that it is slowing down the game. So imagine if VAR could be further improved? Although we’re a long way off watching actual robots playing in sports tournaments rather than human beings, the use of artificial intelligence (AI) in sporting events is already a reality, and it has so much more potential.   Take for example Watson by IBM. It has been used at Wimbledon tennis tournaments for a few years’ now, but this year, Watson – the supercomputer that combines AI and sophisticated analytic software, will be used to change the way fans perceive and enjoy the games. Watson has been taught to better recognise player’s emotion to create automated video highlights – bringing to life the most exciting moments of the championships. The AI engine will use these recognised emotions to deliver tailored video highlights to viewers around the world. With AI, the future VAR would not require a single human referee to monitor the screens. Instead, using deep learning technology, software would automatically detect when a foul has occurred in real-time and alert the referee instantly when a player should be given a red card. There would be no need for constant video replays and discussions – the action in question would have an instant outcome.

“With AI, the future VAR would not require a single human referee to monitor the screens. Instead, using deep learning technology, software would automatically detect when a foul has occurred”

Using video analytics to improve your business

AI video analytics is gaining momentum in all areas of business. It comes with many benefits to businesses, especially those in security and public safety. Data derived from video analytics can produce actionable intelligence to which could help enable law enforcement, and could help even managers and administrators to understand and respond to current situations. For security providers, it saves hours upon hours of time spent trawling through CCTV footage to find that one assailant or key piece of footage to bring a successful conviction. But it doesn’t end there; other industries have used video analytics applications to their advantage, such as education, transportation, banking and financial institutions, casinos, and gaming, and government industries. SCC provides businesses like yours with extensive AI video analytic capabilities, enhancing your business output. Click here to see how video analytics could help you services here