I will build player tracking and performance analysis software for sports footage

Turn raw match or training footage into structured player performance data using AI.

This gig delivers a computer vision system that detects and tracks individual players across video, assigning persistent IDs and extracting performance metrics like speed, distance covered, positioning, and movement heatmaps the kind of data used by coaching staff, scouts, and sports tech platforms.

What I actually build:

• A detection + tracking pipeline (YOLO-class detector + a tracker like ByteTrack/DeepSORT) tuned to your sport and camera setup
• Metric extraction: speed, distance, positional heatmaps, and optional pose-based movement analysis
• Clean data output REST API, JSON export, or webhook, so it plugs into whatever platform you're already using
• A lightweight dashboard if you want a visual layer, not just raw data

Every project starts with the Basic tier a short discovery phase where we look at your actual footage, confirm what's realistically achievable given camera angle/resolution/lighting, and lock in an architecture before any code is written. This avoids the single biggest failure point in sports AI projects: building against footage that was never actually tested.