06 — AI / Computer Vision
Snooker AI
A computer-vision referee for snooker and billiard tables: cameras over each table, object detection on the live stream, and automatic game start, end and scoreboard updates.
Real-time
Table & game detection
- Client
- Snooker AI
- Industry
- Local Business
- Market
- Morocco
- Year
- 2026
Technology
- RTSP IP cameras
- YOLO object detection
- Computer vision
- Live scoreboard
Challenge
In a club, table time is revenue — but staff can't watch every table. Games that start late or end unrecorded distort billing and statistics.
Players also expect a modern experience: live scores and stats rather than chalk and memory.
Solution
We connected RTSP cameras mounted above the tables to a detection pipeline trained to recognise the table and the balls, so the system knows when a frame starts, when it ends and what is happening in between.
Detected events feed a scoreboard and the club's operations data, closing the loop between what happens on the cloth and what is billed.
Implementation
- 01Camera placement and RTSP stream ingestion per table
- 02YOLO-based ball and table detection
- 03Automatic game start and end detection
- 04Live scoreboard and player statistics
- 05Integration with the club's operations app
Result
Tables report their own activity, reducing the manual tracking staff have to do during busy hours.
A foundation for player statistics and automated refereeing features.
Services behind this project
Let's build what's next
Let's build what's next.
Have a project, product or ambitious idea? Tell us where you want to go.