Develop a Custom Soccer Video Analysis Tool Using AI -- 2

I am looking for a developer to build a custom video analysis tool for soccer matches. The goal is to process full match footage and automatically tag events such as goals, passes, player tracking, possession analysis, average player positions, xG calculations, and more.

Key Deliverables:

1) Object Detection and Tracking:

Use YOLO (or similar) for player and ball detection.

Integrate DeepSort or a comparable tracking library for consistent player tracking across frames.

2) Event Detection:

Automatically detect events such as goals, passes, shots, etc.

Implement logic to analyze possession time and calculate xG.

3) Output Data:

Generate reports with average player positions, event summaries, and possession breakdowns.

Include visual overlays on the video showing tracked objects and detected events.

4) Video Processing Pipeline:

Efficiently process full match footage (90+ minutes).

Ensure compatibility with common video formats.

5) Code Documentation:

Provide clean, modular, and well-documented code for future updates or enhancements.

I want to be able to upload footage captured by a Veo camera for example, and use this to generate stats for analysis. Please attach your estimated timeline and cost for the project. Any suggestions or improvements you can propose for the requirements listed. I will only be replying to people with direct experience and people who can provide evidence of similar projects completed in the past.

Coding Machine Learning (ML) Artificial Intelligence Sports

Project ID: #39016669

About the project