Soccer Player Tracking System

Google Drive Submission Package

Drive Folder Structure:

PLAYER_TRACKING/
├── README.md               (This file)
├── Technical_Report.pdf
├──  Code/
│   ├── main.py               (Main program)
│   ├── detector.py           (YOLO detection)
│   └── tracker.py            (Re-ID logic)
├──  Models/
│   └── best.pt            (Pre-trained weights)
├──  data/
│   ├── input_videos.mp4             (Sample input)
├── output/            (Processed result)
└──  requirements.txt       (Dependencies)

Quick Start Guide

1. Download Files

1. Download the entire folder from Google Drive
2. Preserve the folder structure exactly as shown above

2. Install Requirements

pip install -r requirements.txt

• Pro Tip: Use Python 3.8+ and a virtual environment*

3. Run the Program

python Code/main.py

▶ Expected Output:

• Processed video saved to Videos/output.mp4
• Terminal will show real-time progress:

  Processing Frame 45: 12 players tracked (1.3s/frame)
  

Detailed Setup

For Custom Videos:

1. Replace Videos/input.mp4 with your video
2. Edit these paths in Code/main.py:

   INPUT_VIDEO = "Videos/your_video.mp4"
   OUTPUT_VIDEO = "Videos/custom_output.mp4" 

How It Works

Core Pipeline:

1. Frame Extraction → Reads video at 30 FPS
2. Player Detection → YOLOv11 model
3. Feature Storage → Saves jersey color/position
4. Re-Identification → Matches players across frames

Live tracking example

Need Help?

Common Issues:

• "Missing model weights" → Ensure yolov11.pt is in ⚙️ Models/
• Slow performance → Try reducing resolution in detector.py
• Video playback issues → Use VLC Player

Contact Support: [email protected] (mailto:)

---