Project challenges
In its classical form, this process is based on browsing a video from the latest match by coach staff, which gives the opportunity to evaluate every player. This is a time- consuming process because of the number of players and match duration. Additionally this process is subject to human error risk.
Technological solutions
The system is based on a match video, recorded in real time or loaded from a previously made video. Next the material is processed by the Machine Learning module which identifies key elements, like pitch surface, ball and footballers. The system enables measuring metrics and statistics for each player individually and for all team. Predicted values are designated by the special predict module based on historical data.
Python
TensorFlow