AI can significantly cut down on the Amount

of time needed to determine whether a Premier League football game is offside.

 

 

 

 

 

With the commencement of the English Premier League (EPL) football season, additional modifications are being made to the VAR (video assisted referee) system. In an effort to lower the amount of erroneous refereeing decisions, such as offsides, VAR was introduced to the English Premier League in 2019. Despite this, its use continues to draw criticism.

Based in Stockley Park, west London, referees and other officials review video replays of judgments made during officiating sessions in real time. The referee on the field is then given feedback from these officials and the footage of the crucial moment to help guarantee that their decisions are accurate.

The Premier League asserts that, compared to 82% prior to the implementation of VAR, the right judgment is now made 96% of the time. After VAR was implemented, academic studies have also shown that decision-making improved. Approximately 1,300 EPL match incidents were examined in the previous season.

 

The length of time it takes to complete a review is one of the main complaints of VAR, and the EPL is going to start using artificial intelligence (AI) to remedy this.

 

Semi-automated offside technology (SAOT) uses artificial intelligence (AI) to track players at high speeds and in great detail. It was previously utilized at the 2022 World Cup in Qatar and the 2024 Euros.

SAOT is thought to be able to cut offside decision times by as much as 31 seconds. Additionally, it gives viewers on TV who are seeing these offside infractions a clearer picture. Instead of manually drawing two lines one each for blue and green or red across a picture of the pitch, SAOT projects a virtual vertical curtain that indicates precisely where portion of the player’s body is offside.

 

Computer vision systems, the foundation of the technology, are capable of tracking up to 10,000 surface body points on each player as well as the ball. The data business that won the EPL’s SAOT contract, Genius Sports, claims that it can track players at a rate of 200 updates per second.

Modern artificial intelligence (AI) and deep-learning algorithms technology that mimics how the human brain makes decisions make it possible to track objects with such high detail and speed. It is capable of processing enormous volumes of video footage from the dozens of cameras placed at every EPL football stadium.

 

How it functions

 

One kind of deep learning algorithm that is especially meant to process and understand the contents of images and videos is called a convolutional neural network (CNN). They are essential to SAOT’s functioning.

CNNs use a mechanism known as a multi-stage hierarchical process to evaluate images.

Basic features from the image, such as edges, blobs, and corners, are extracted by layers early in the processing pipeline, and these features are aggregated and assembled into bigger, conceptually more relevant categories by layers later in the pipeline.

 

The CNNs are customized and trained to generate an estimate of the spatial coordinates of each point on a player’s body and the ball for each streamed image when they are used for surface point tracking, like in the case of Genius Sports’ computer vision technology.

However, before the AI system can become more proficient at these tasks, it must first be “trained” that is, given several instances of what it should be searching for. The computer system must endure a taxing process during this mostly automatic training phase before it can produce real-time forecasts for SAOT.

 

VAR might never be flawless.

 

The use of artificial intelligence (AI) in research and engineering is expanding the realm of sports data analytics in many ways. One notable example of this is the use of SAOT in professional football matches.

After EPL executives are satisfied that SAOT is up to par, the introduction is anticipated to begin in October 2024. Tony Scholes, the chief football officer of the Premier League, stated that VAR might “never be perfect.” But SAOT’s arrival will undoubtedly bring it closer, which could aid in the technology’s increased acceptance by the majority of fans.

 

If VAR is improved, many supporters say they are in favor of it. Apart from the duration of decision-making, other critiques include a decrease in fan satisfaction, a reduction in the impromptu nature of goal celebrations, and an impression of fussiness over insignificant choices.

Wolverhampton Wanderers even proposed ending the Premier League’s usage of VAR during a meeting of clubs at the end of the 2022–2023 season, however Wolves lost this vote 19–1.

As supplied by The Conversation