👁️ Computer Vision

Hand Gesture Home Automation

Vision-Based Smart Home Control

1 month Solo Project Completed

Overview

Control lights and fans via intuitive hand gestures using MediaPipe and IoT integration with 95% recognition accuracy.

The Problem

Physical switches and voice control aren't always convenient or accessible for all users.

The Solution

Vision-based interface using MediaPipe hand tracking to recognize gestures (thumbs up, peace sign, fist, etc.) and control IoT devices via MQTT protocol.

Results & Impact

95% Gesture Accuracy Across 6 gestures
<200ms Response Time Gesture to action
$35 Cost Using standard webcam

Key Impact

  • Accessible control for mobility-impaired users
  • No additional hardware beyond webcam required
  • Extensible to other smart home devices

Technologies Used

PythonMediaPipeOpenCVMQTTArduinoESP32