We turned a game everyone knows into a live AI showcase for computer vision and GenAI

We were challenged to apply computer vision to a game almost everyone has played: foosball. In partnership with Ten Twenty One, we created ChutAI: a system that tracks the match in real time, predicts who is most likely to win and even generates live commentary using GenAI. What started as a playful experiment is now showcased at major events to inspire young people and demonstrate how far these technologies can go beyond traditional business use cases.
Implementation time:
2 months
foosball games covered
events with ChutAI
people reached
Foosball is simple on the surface, yet perfect to highlight the challenges of running real-time computer vision algorithms on a very fast paced setup
Foosball is simple on the surface, yet perfect to highlight the challenges of running real-time computer vision algorithms on a very fast paced setup
The goal was not only to analyse the game but to make the experience fun, interactive, and accessible to everyone
We mounted a vertical camera above the foosball table and built a system that captures and analyses every moment of the match. Using computer vision algorithms, we track the ball and players in real-time and calculate statistics such as goals, passes, shot speed, ball possession, off-target shots, and touches. With this live data, a probability model estimates which team is most likely to win at any point.
The dashboard displays dynamic graphs and real-time metrics for players and spectators, turning each match into a data-driven experience. To make it even more entertaining, we experimented with GenAI to generate live commentary, giving every game the energy of a sports broadcast.
ChutAI premiered at Portugal Google Cloud Day 2024 and is now used at company events and tech fairs to show young people the creative side of AI and the possibilities of applying computer vision beyond traditional business problems.
Meet the team of this project

Tiago Mota
Senior Machine Learning Engineer

Patrick Caetano
Data Engineer
ChutAI quickly became the star of every event where it was showcased. Players were surprised to see their moves translated into instant metrics, while spectators gathered around the screen cheering at the live stats and laughing at the AI-generated commentary. What began as a technical demonstration became an experience that blended play, competition, and technology in a way people had never seen before.
By analysing every shot, predicting win probability in real-time, and delivering engaging commentary, the system brought the excitement of professional sports analytics to a simple foosball table, proving that AI can be both powerful and fun.
The technology behind ChutAI goes far beyond foosball. In entertainment, it can power real-time sports metrics, audience-engagement tools, and virtual commentators for any sport or event. In industry, the same computer-vision foundations can support production optimisation, predictive maintenance, quality assurance, and smarter logistics. In healthcare, these models can enable remote monitoring, automated interactions, and personalized recommendations, opening the door to safer and more accessible patient support.
Meet part of the team of this project

Tiago Mota
Senior Machine Learning Engineer

Patrick Caetano
Data Engineer
Partners

Awards

.png)