UEFA Euro 2024 relies on the latest data tracking technology to enable accurate and quick decisions in offside and goal situations. Let's take a closer look at the techniques used.
Football is becoming more and more a data-driven game, as happened at the 2024 European Championship in Germany.
Photo: Panther Media/Federico Caputo
From a purely formal perspective, old Sepp Herberger sayings like “a game lasts 90 minutes” (often much longer today) or “the ball is round” still apply. In fact, the game of football in 1954 doesn't have much in common with the game of today. Everything has become much faster and sportier, balls and shoes are made of light plastic and no longer made of heavy leather, which becomes heavier when it rains. This is on muddy ground, while today's green grass looks more like a carpet.
In the 1954 World Cup, fans were happy to see snippets of matches on television. Seeing the entire final on screen was a huge buzz. On the other hand, at the European Championships in 2024, spectators will be treated to all kinds of artistic prowess. The European Football Championship represents not only exciting matches and exciting emotions, but also a technical revolution on the pitch. State-of-the-art data capture technologies ensure that every in-game action is accurately recorded and analyzed. But how exactly does this technology work and what impact does it have on referees' decisions and viewers' experience?
High-tech ball “Love Football”
The Love Football ball is more than just a football. Inside is an advanced sensor core consisting of a wide-bandwidth (UWB) sensor and a motion sensor (IMU). These sensors record the ball's position and movement up to 500 times per second. The data is sent to a Local Positioning System (LPS) located around the stadium, which calculates locations with an accuracy of up to 10 centimetres.
The ball sensor core is integrated in such a way that the roll and flight characteristics are not affected. The UWB sensor sends its location data 100 times per second, while the IMU sensor equipped with acceleration and tilt sensors sends data 500 times per second. This technology ensures that the exact time of ball contact and ball position are known at all times. By the way, the commercially available balls labeled “Football Love” are not equipped with high technology.
Pitch as a 3D scanner
In order to fully capture what is happening on the field, up to 32 high-performance cameras have been installed in the stadiums. These cameras are strategically placed under the roof of the stadium and monitor every movement of the players. In addition, there are between 12 and 24 antennas around the pitch that receive signals from sensors built into the ball. Cameras and antennas work together to create a full 3D model of the game in real time.
Cameras capture every movement players make from different angles and use algorithms to create detailed 3D models of players' skeletons. This technology, known as SkeleTrack, makes it possible to track 29 key points on players' bodies and accurately visualize movements. This technology has been developed and tested in collaboration with leading research institutions such as the MIT Sports Lab and ETH Zurich.
Real-time data analysis and visualization
The data collected is used not only to support the judges, but also for real-time perceptions of the viewers. These visualizations include automatically generated lines and marks on players' relevant body parts. This allows quick and transparent assessment of controversial situations such as offside situations or goal decisions.
By combining sensor data with AI-controlled camera systems, the game can be analyzed in real time and visualized in a way that is understandable to the viewer. This helps ensure that the audience on the field and in front of the screens can understand the referees' decisions.
This combination of spherical sensor and camera technology was used for the first time at the 2022 World Cup in Qatar. One of the most notable events is the match between Portugal and Uruguay. The sensor ball proved that it was not Cristiano Ronaldo, but Bruno Fernandes who scored the decisive goal. The sensor data clearly showed that Ronaldo did not touch the ball as no vibrations were recorded.
Semi-Automatic Stealth Technology (SAOT)
One of the biggest challenges in football broadcasting is the offside decision. Semi-automatic stealth technology (SAOT) is used here. This system uses machine-enhanced video images to help assistant referees in the Catacombs quickly and accurately assess offside situations. However, this technology can only detect potentially suspicious game situations and hand them over to a human referee for evaluation.
The SAOT system dramatically shortens the time needed to make an infiltration decision. While the classic VAR system takes up to 70 seconds, SAOT can make a decision in about 15 to 25 seconds. Despite the rapid processing of data, the human factor remains key, as technology recognizes offside situations but cannot interpret all the details of the rules.
Take, for example, what is called “passive stealth.” It is often obvious that the offside player is not interfering in the game, obstructing the opponent or blocking the goalkeeper's vision. But sometimes you also need a trained eye of judgment to evaluate this. Even here there is a lot of freedom of action, so there are no clear opinions about whether a player is passively offside or not.
Details and technical challenges
Without SAOT technology, the assistant referee had to manually search for crucial frames and draw lines to determine the positions of players and the ball. Critics complained that the purely video-based system was not accurate enough because the 50 frames per second refresh rate was insufficient. Additionally, studies have shown that people timed the playback an average of 132 milliseconds too late.
Players, on the other hand, do not have sensors implanted or attached, but are followed at their every step using optical tracking methods and measured down to the tips of their shoes. The twelve cameras located on the roof of the stadium are used for this purpose, and they record the scene from different angles. Object recognition algorithms use images to measure 29 points of interest on players' bodies and use them to create artificial skeletons to be able to determine positions with centimeter accuracy and visualize movements.
Hawkeye and SkeleTrack
The limb detection technology was developed by Hawk-Eye and popularized as SkeleTrack. This technology allows tracking 29 important points of the players' body and accurately visualizing movements. This technology was developed and tested in collaboration with leading research institutions such as the MIT Sports Lab and ETH Zurich.
Artificial intelligence plays a crucial role in data processing. Deep convolutional neural networks (Deep CNN) recognize patterns in video images and determine the location of players' limbs. The open source framework OpenPose, developed by scientists at the University of California and others, allows tracking multiple people in real time and analyzing their movements in detail.
Technology limits
Despite all the progress, there are also limits. Edge detection can fail in turbulent game situations, for example when players are covering each other. In addition, unfavorable lighting conditions and distortions in the training data can affect the accuracy of predictions.
Although data is carefully collected and analyzed, technology cannot accurately evaluate every situation in the game. Fouls can happen, especially in dynamic, contentious game situations in which players operate on the edge of the offside rule. Inadequate lighting conditions or obstacles imposed by other players can affect detection accuracy. Therefore, the final decision always rests with human judgment.
“Certified tv guru. Reader. Professional writer. Avid introvert. Extreme pop culture buff.”
More Stories
Remotely controlled cargo ships coming soon on the Elbe Canal?
Siemens technology makes Baden Canton Hospital smart
Discovering an ancient Mayan city – what do the rainforests hide?