I will develop object detection, player ball tracking system for football analytics


About this gig
Want to automatically analyze football games with AI?
I will develop a computer visionbased tracking system that detects and follows all players, referees, and the ball throughout a match turning your video into structured data and clear performance metrics.
Using YOLOv8, DeepSORT, OpenCV, and PyTorch, this system maintains consistent IDs, computes player distances, and outputs match analytics in CSV or dashboard form ready for tactical review or performance improvement.
What You will Get :
- Player, ball, and referee detection
- Auto player ID assignment across frames
- Distance, speed, possession, and coverage metrics
- Optional heatmaps and visual overlays
- Structured outputs: CSV / JSON / XLSX
- Integration-ready codebase for future expansion
Tech Stack: YOLOv8, DeepSORT, OpenCV, PyTorch, TensorRT
Output Formats: CSV, JSON, XLSX, or video overlay (MP4)
Why Me:
- Expertise in football and multisport CV systems
- Accurate tracking even under occlusion or camera motion
- Clean, optimized Python scripts with documentation
- MVP-ready within weeks
Lets build the foundation of your AI-powered football analytics system today.
Get to know Kim L
sports computer vision and AI analytics developer
- FromUnited Kingdom
- Member sinceNov 2025
Languages
English, Spanish
FAQ
Q: Can this work for full 90-minute games?
A: Yes, the model can process long videos in segments or full matches depending on hardware.
Q: Does this support multiple cameras?
A: The current version works per camera; multi-view tracking is available on request.
Q: Can I customize the stats?
A: Absolutely — I can tailor metrics to your specific KPIs or team performance needs.
Q: Will I receive the Python source code?
A: Yes, included for Standard and Premium packages.

