Skip to content

Gesture-Based Drone Control

A real-time computer-vision system that eliminates the physical controller entirely. Hand gestures detected through a live camera feed, classified by a dual rule-based and ML engine, translated directly into drone flight commands — with safety fail-safes built in from the ground up.

Get Started   GitHub ↗


Core Intelligence

Dual Recognition Engine

MediaPipe 21-point landmark detection feeding a rule-based classifier and TFLite ML model.

Real-Time Pipeline

Millisecond Latency

Live camera → gesture classification → command translation → drone. FastAPI WebSocket backend.

Safety First

Built-In Fail-Safes

Hover on tracking loss, emergency stop gesture, idle detection auto-land — active by default.