FITNESS VIRTUAL
ASSISTANT

a raspberry pi that watches your form

A Raspberry Pi-based project that tracks fitness progress by analyzing body movements in real time — things like bicep curls and shoulder presses — using pose estimation instead of a phone camera app.

Stack

Raspberry Pi Python MediaPipe Sense HAT Google Coral TPU

Project features

> MediaPipe integration — detects body landmarks and tracks joint angles for real-time pose estimation.

> Exercise tracking — counts bicep curls and shoulder presses by analyzing joint angles.

> Speech recognition — voice commands and real-time audio feedback for hands-free use.

> Sense HAT integration — captures temperature, humidity, and pressure to enrich workout feedback.

> Real-time pose estimation — accelerated with a Google Coral TPU for fast, responsive feedback.

> Leadership — led full-stack development and collaborated with a cross-functional team to deliver the prototype on schedule.

Status

Shipped ✓