Soccer Analysis System

Professional Computer Vision & Machine Learning Project

👋 About This Project

This repository contains a comprehensive soccer analysis system that demonstrates advanced computer vision and machine learning techniques. The system performs real-time player tracking, team classification, tactical analysis, and database integration with professional-grade data export capabilities.

🎯 Project Overview

The Soccer Analysis System processes soccer videos to extract detailed player tracking data, perform tactical analysis, and export comprehensive datasets. It features a split-screen interface with real-time video analysis and tactical board visualization.

🎯 Technical Challenges Solved

This project addresses several complex computer vision challenges in sports analytics:

• Real-time Player Tracking: Implemented robust player tracking using YOLO and ByteTrack to maintain consistent player identification across frames
• Team Classification: Developed automated team classification using SigLIP feature extraction and machine learning clustering
• Multi-coordinate Transformation: Created coordinate transformation pipeline from video pixels to real-world pitch coordinates
• Database Integration: Built comprehensive PostgreSQL integration with mock database fallback for testing
• Data Export Pipeline: Implemented complete ETL pipeline with CSV/JSON/TXT export formats
• Professional UI: Developed split-screen interface with tactical board visualization and real-time progress tracking

💻 Installation

Install the system in a Python>=3.8 environment.

# Clone the repository
git clone https://github.com/ashok-sravanam/sports-optimized-.git
cd sports-optimized/examples/soccer

# Install dependencies
pip3 install -r requirements.txt
pip3 install -r tactical_requirements.txt

# Download models and sample video
chmod +x setup.sh
./setup.sh

🚀 Quick Start

# Run the analysis system
python3 test_bug_fixes.py \
    --source_video_path "video_outputs/psgVSliv.mov" \
    --target_video_path "video_outputs/analysis_output.mp4" \
    --max_frames 500 \
    --device cpu

🎯 Features

• Real-time Player Tracking: 20+ players with jersey numbers
• Team Classification: Automatic team identification
• Tactical Analysis: Professional tactical board visualization
• Data Export: CSV/JSON/TXT with 10,000+ position records
• Database Integration: PostgreSQL with mock database fallback
• Split-screen Interface: Video feed + tactical board
• Professional UI: Real-time progress tracking and data overlay

📊 Output

The system generates:

• Video: Split-screen analysis with tactical board
• Data: Complete CSV export with all tracking data
• Statistics: JSON summary with player metrics
• Reports: Human-readable analysis reports
• Documentation: Comprehensive data column guides

🏆 Technical Achievements

• Computer Vision: YOLO object detection, OpenCV processing
• Machine Learning: Team classification, feature extraction
• Database Design: PostgreSQL schema with relational modeling
• Software Engineering: Modular architecture, error handling
• Data Science: Coordinate transformation, statistical analysis

---

Technologies: Python, OpenCV, YOLO, PostgreSQL, scikit-learn
Repository: https://github.com/ashok-sravanam/sports-optimized-