World Cup refereeing decisions are debated decades later.
From whether the ball crossed the finish line in 1966, to Diego Maradona’s ‘hand of God’ two decades later, to some of the decisions made by the video assistant referee at Russia 2018, any mistakes the referee sees will be scrutinized by fans years later.
Rulers need all the help they can get, and they could be about to get a helping hand from AI.
Over the past few years, FIFA has been testing the use of offside party tracking technology, which uses artificial intelligence as well as a series of cameras around the field to track players’ limbs and instantly create virtual offside lines for referees. The technology has been used so far in the FIFA Club World Cup and the FIFA Arab Cup, and FIFA expects to do so Used in Qatar 2022.
Given all the issues with VAR technology, there may be concerns about the use of AI in football.
But Dr. Patrick Lucy, chief scientist at Sports data company Stats Perform He says FIFA’s approach to member tracking is the right one.
He says that instead of fully automated decision making, artificial intelligence is used to give accurate measurements and create offside lines that previously had to be drawn manually. This removes human error, but humans are still very knowledgeable.
The referee or assistant referee can look at the image created by the AI and instantly know if a player is offside, but the human official can also use their own judgment to decide if that player is interfering with play or if there is some other reason. Why should a target be given or not.
Dr. Lacey explains that this approach combines “getting humans to do what they do really well and making computers do what they do really well.” That is why the head of the FIFA Referees Committee Pierluigi Collina said it was not a “stealth robot”.
Since the manual work is done by a computer and not a human, the decision will not be as instantaneous as the target line technology, but it should be much faster than the current VAR technology, which will hopefully eliminate some of the frustration caused by lengthy VAR decisions. , as well as improving the general decision-making of the rulers.
Dr. Lucy points out that player tracking technology is not new, it has been used in basketball for over twenty years. But camera technology has improved with 4K and 8K cameras with higher pixel density, and AI can now do things that weren’t possible before.
With stronger artificial intelligence and improvements in player tracking technology, Dr. Lucy says that within a year manually collected football data and metrics can be captured directly from a live football TV broadcast without the need for other cameras. in the place.
Although broadcast TV does not provide the quality of live member tracking technology needed to make accurate management decisions, AI will soon be able to use broadcast TV to generate real-time statistics, including projected goals and the estimated position of players off-screen. This is done by using a fingerprint of player tracking data from the past 20 years along with machine learning algorithms to estimate the whereabouts of missing players.
The technology can then be used to broadcast television historical matches, collecting data from previous World Cups, allowing statisticians to analyze data on expected goals from England’s golden generation or Diego Maradona.
One of these advanced stats that AI can generate is called “shadows.” This AI is used to determine if players are in the position they are supposed to be in certain situations. Clubs can use it to correct errors, such as showing if a winger is in the wrong position when counterattacking, or to find weaknesses in their opponents, and see what players are likely to do in a given situation.
Since AI has made it possible to generate such stats from a TV show, this will make advanced stats available at a much broader level of the game, which can be more impactful than member tracking technology.