AirPointer is a real-time hand gesture-based cursor and browser navigation system. It leverages MediaPipe to track hand landmarks and uses PyAutoGUI to perform seamless navigation and cursor control without the need for physical input devices. This system enables hands-free interaction, making it useful for accessibility, touchless control, and immersive experiences.
- Real-time Hand Tracking: Uses MediaPipe to detect and track hand landmarks.
- Cursor Control: Navigate the cursor using hand gestures.
- Click & Scroll: Perform left-click, right-click, and scrolling actions with specific hand gestures.
- Browser Navigation: Seamlessly switch between tabs, refresh, open, and close tabs with intuitive gestures.
- Mode Switching: Toggle between cursor movement and browser navigation mode using a thumbs-up gesture.
- Hands-Free Interaction: Enables touchless control for accessibility and convenience.
- Index Finger Left: Navigate to the previous tab.
- Index Finger Right: Navigate to the next tab.
- Open Palm: Open a new tab.
- Closed Palm: Close the current tab.
- Victory Sign: Refresh the current tab.
- Three Fingers Up: Scroll up.
- Hang Loose Gesture: Scroll down.
- Thumbs-Up Gesture: Toggle between cursor movement and browser navigation mode.
- Index Finger Movement: Move the cursor.
- Okay Sign: Left-click.
- Double-Finger Okay Sign: Right-click.
- Double Tap Okay Sign: Double-click.
- Python
- MediaPipe (for hand tracking)
- OpenCV (for video processing)
- PyAutoGUI (for controlling the mouse and browser navigation)
- NumPy (for efficient computations)
- Clone this repository:
git clone https://github.com/Sania-hasann/AirPointer.git cd AirPointer - Install the required dependencies:
pip install mediapipe opencv-python pyautogui numpy
- Run the script:
python airpointer.py
- Accessibility: Provides an alternative input method for individuals with limited mobility.
- Touchless Navigation: Useful in public kiosks, smart homes, and hygiene-sensitive environments.
- Immersive Interaction: Enhances user experience in gaming, presentations, and virtual environments.
- Automation & Productivity: Allows fast tab switching, scrolling, and navigation without using a keyboard or mouse.
- Assign easy hand gesture for navigation.
- Enhance gesture recognition accuracy.
- Support for multiple applications beyond browsers.
- Customizable gesture mappings.
- Multi-hand gesture recognition for extended functionalities.
Feel free to fork this repository, raise issues, and submit pull requests!
This project is licensed under the MIT License.