Overview
A Final Year Project for VIT, this project demonstrates a novel way to control a computer interface using hand gestures and voice commands. It leverages MediaPipe for real-time gesture recognition, integrates a custom voice assistant named Krishna, and incorporates generative AI (using Google Gemini) for enhanced query responses.
Key Features
- Cursor Control: Move the mouse pointer using hand movements.
- Click Operations: Perform left, right, and double clicks with specific finger gestures.
- System Controls: Adjust system volume and screen brightness using pinch gestures.
- Voice Assistant (Krishna): Launch gesture recognition, Google search, file navigation, and enter Divine Mode.
- Divine Mode (GenAI): Leverage Google Gemini Flash for generative AI responses including screenshot analysis.
Tech Stack
- Python 3.9
- MediaPipe
- OpenCV
- Google Gemini API
- PyAutoGUI
- PyCAW