TennisFit

TennisFit explores a connection between how someone plays and the equipment they choose. The goal is to use a tennis video to build an understandable picture of a player’s style, then suggest racquet and string characteristics that may suit it.

The current Python prototype connects video processing, player and ball tracking, court positioning, and pose-based motion features to a rule-based playing-style profile. A local Streamlit demo brings the analysis and recommendations together.

The recommendation layer combines that profile with preferences for power, control, spin, and comfort. It suggests equipment attributes, such as racquet head size and string type, with explanations of the rules behind each suggestion.

This is an early prototype. Several tracking components still rely on heuristics, and the equipment recommendations have not yet been validated against player feedback. The next steps are to improve tracking on real tennis footage, expand the equipment data, and collect feedback to evaluate and refine the recommendations.