ai goalkeeper trainer

This computer vision project implements an interactive hand-tracking game that challenges users to quickly "hit" on-screen targets. Leveraging the metaidigitcv library, which is built upon MediaPipe, the system accurately detects and tracks hand landmarks in real-time.

A key feature is the quadratic mathematical function employed to estimate the real-world distance of the user's hand from the camera based on the pixel-width of the detected hand. This allows the game to determine when the hand is close enough to interact with a target.

The game displays a grid of targets, where a new target appears randomly after a successful "hit." Visual feedback, including color changes and an overlay on the hand, guides the user, while a score and timer track performance. This project demonstrates practical applications of hand tracking, distance estimation using non-linear functions, and basic game logic within a computer vision framework.