This project utilizes YOLO, a state-of-the-art object detection algorithm, to accurately identify and track players, referees, and the football in video footage. YOLOv5 has been fine-tuned on a custom dataset to optimize detection performance for players, referees, and football, ensuring precise and reliable results. Players detected in the video are assigned to respective teams with the help of pixel segmentation, enabling team-based analysis of player movements and interactions. Additionally, it calculates the control of the ball under each team as a percentage, providing insights into ball possession and team dominance during gameplay and, measures the speed and total distance covered by players.
pramodiperera/Football_Analysis
YOLO-based advanced football analysis: Object detection, pixel segmentation, and object tracking for comprehensive performance insight
Jupyter Notebook