The Grassroot Football Analysis project aims to develop a comprehensive system for analyzing grassroots football match videos using advanced AI techniques and Big Data infrastructure. By leveraging state-of-the-art player detection, tracking, and performance analysis tools, this initiative seeks to support training strategies, improve player development, gain tactical insights, and enhance fan engagement.
This project integrates:
- Player tracking, speed estimation, and ball movement analysis
- Automatic team identification using clustering
- Big Data management for large-scale football video datasets
- Tactical analysis to improve performance and game strategies
You can access the report here: 📄 Project Report
✅ Player Detection & Tracking: Identifies and follows players using YOLOv5 & YOLOv8
✅ Camera Motion Estimation: Adjusts player positions for accurate tracking
✅ Field View Transformation: Corrects perspective distortions for an ideal tactical view
✅ Ball Position Interpolation: Estimates ball location when not directly visible
✅ Player Speed & Distance Estimation: Analyzes movement patterns and fitness levels
✅ Team Classification via Clustering: Uses K-Means clustering to automatically group players by team colors
✅ Ball Possession Analysis: Determines which team controls the ball
✅ Big Data Video Management: Supports large grassroots match video datasets for long-term tracking and analysis
git clone https://github.com/Amiche02/Advanced-Football-Video-Analysis.gitpip install -r requirements.txtwget https://www.dropbox.com/scl/fo/hybscdrucozk29pbda15u/AN3zedKh1YYlkVtnXFH13Vk?rlkey=8rr5owurn3mu6pfarofh7tctq&st=s96ogmos&dl=0 -P models/1️⃣ Place input videos in the input_videos/ folder.
2️⃣ Run the main script to analyze the videos:
python main.py3️⃣ Output videos will be saved in the output_videos/ folder.
Detecting and tracking players using YOLO models:
Automatically assigning players to teams based on jersey color:
We welcome contributions! Feel free to:
- Open an issue for bugs or feature requests.
- Submit a pull request if you improve or optimize the code.
The Grassroot Football Analysis project represents a significant step in applying AI and Big Data to grassroots football. By analyzing match videos, we aim to refine training methods, enhance player development, and provide deeper insights into team strategies.
As we continue improving, we encourage the football and AI communities to join us—whether through data contributions, technical enhancements, or sharing insights. Together, we can unlock new potential in grassroots football analytics! ⚽📊🚀
🔔 Stay tuned for more updates!