AI is Helping Referees Call Offside in FIFA World Cup 2022

by Meghmala December 2, 2022

FIFA 2022

Referees are assisted by SAOT-based AI to call offsides at the 2022 FIFA World Cup.

At this year’s World Cup, FIFA is using new artificial intelligence to help referees make calls. Twelve cameras installed on the roof of the stadium work as a semi-automated offside technology (SAOT) system, which follows the ball and the movements of each player. Qatar’s official 2022 World Cup ball, Al Rihla, which means “journey” in Arabic, has an attached sensor that enables SAOT to compare the exact moment it was kicked with the position of the team’s last defender and the other team’s forward. . SAOT tracks and detects players and the ball using artificial intelligence, calculating their location 50 times per second. The game officials are notified whenever the SAOT is found offside. The referee, who has the final say, is notified. The system is described as “semi-automated” as a result. This level of accuracy is important in close calls where referees struggle to quickly identify offside. In this case, the goal or the result of the whole game, can depend from time to time.

Lire Aussi :  The Global Artificial Intelligence (AI) Market is expected to grow by $125.3 bn during 2022-2027, accelerating at a CAGR of 22.26% during the forecast period

A computer’s hypothesis will be tested against known factors using a classification method when there is enough information to make educated decisions. Typically, large video datasets of objects that have been seen by humans – in this case, soccer players on the field – are used to train artificial intelligence systems like SAOT. Artificial intelligence captures the appearance of players in this way. Following extensive training, this technology can detect and track players with ease and speed.

Football games often use video assistant referee (VAR) technology. It takes much longer than SAOT to detect offside—70 seconds. The officials had to decide the correct time for the kick and mark the offside line without the help of VAR technology. They only need to check the offside recommended by the system and SAOT. According to FIFA’s website, this new process “happens in seconds and means that offside decisions can be made quickly and accurately.” If the referee agrees with SAOT’s recommendation, the system will generate a 3D cartoon for the offline broadcast to be shown to the spectators on the big screen in the stadium.

Lire Aussi :  The World's Funkiest Body Pose Tracker

To solve this problem, data scientists have developed various methods. Convolutional neural network is one of them (CNN). CNNs work by detecting objects layer by layer. When you press an object, you find that it is soft in some places and hard in others. Your perception of this object changes because of this movement: You now have enough knowledge to realize that it has both soft and hard properties. The first “layer” of discovery will be represented by this information. This will be called “convolution” in CNNs. Following the initial layer identification, you will ask additional questions about the object’s texture, size, and shape. Every time one of these questions is answered, a new layer emerges, deepening your understanding of the situation. This is very similar to how CNN works.

Lire Aussi :  Tulip's Point of Sale supports Tap to Pay on iPhone with Adyen

According to FIFA, the SAOT system is currently “the most accurate support system available to video game administrators” after three years of testing. A computer’s hypothesis will be tested against known factors using the classification method when there is enough information to make educated decisions. Typically, large video datasets of objects that have been seen by humans – in this case, soccer players on the field – are used to train artificial intelligence systems like SAOT. Artificial intelligence captures the appearance of players in this way. Following extensive training, this technology can detect and track players with ease and speed.

Share This Article

Do something to share

About the author

More information about the author

Source

Leave a Reply

Your email address will not be published.

Related Articles

Back to top button