A hand‑gesture control system that lets you navigate slides, web pages and applications using your webcam, without touching the keyboard or mouse.
This project uses Python (OpenCV + MediaPipe) for gesture detection and sends relevant keyboard events (next/previous/zoom) to control active applications.
- 🟨 Double‑tap on right side → Next page / slide
- 🟥 Double‑tap on left side → Previous page / slide
- 🤏 Pinch close (thumb + index) → Zoom in
- 🤏 Pinch open (thumb + index) → Zoom out
- Works on PowerPoint, Google Slides, Chrome, PDF, and other apps
WhatsApp.Video.2026-01-31.at.6.10.02.PM.mp4
The Python client uses your webcam to detect hand landmarks in real time using the MediaPipe library. It tracks the position of your index finger and detects gestures like double taps and pinches. These gestures are then mapped to keyboard shortcuts using PyAutoGUI.
Gesture detection runs at video speed so interaction feels smooth and natural without touching the keyboard or mouse.
GestureAnaltucsPlatform/ │ ├── backend/ # Node.js backend for analytics │ ├── server.js │ ├── package.json │ └── ...other backend files │ ├── python-client/ # Python gesture client │ ├── hand_double_click.py │ ├── hand_navigation_zoom.py │ ├── gesture_detector.py │ ├── hand_test.py │ ├── mouse_gesture_control.py │ ├── requirements.txt │ └── .venv/ # Virtual environment (ignored) │ └── README.md # This file