Gesture Recognition System

An AI-powered virtual mouse system enabling hands-free computer control through intuitive hand gestures.

✨ Features

• 🖐️ Gesture-Based Control: Control cursor movements and execute clicks using hand gestures
• ♿ Accessibility-Focused: Designed to enhance computer accessibility for specially-abled individuals
• ⚡ Real-Time Processing: Utilizes MediaPipe for efficient hand tracking and OpenCV for gesture recognition
• 🎯 High Accuracy: Achieves 92-97% gesture recognition accuracy in controlled environments

🛠️ Technologies Used

• Python 3.8.5
• MediaPipe (Hand tracking)
• OpenCV (Computer vision)
• PyAutoGUI (Mouse control automation)

📦 Installation

Step 1: Create a virtual environment

bash conda create --name gest python=3.8.5 conda activate gest

Step 2: Install dependencies

bash pip install -r requirements.txt

Step 3: Run the application

bash python Gesture_Controller.py

Gesture Controls

Gesture	Action
Move hand	Move cursor across the screen
Index finger extended	Left click
Middle finger extended	Right click
Quick double index motion	Double click
Closed fist	Drag and drop / Scroll
Open palm (5 fingers)	Cancel or reset current action
Pinch (Thumb + Index Up/Down)	Increase or decrease system volume
Pinch (Thumb + Index Right)	Increase or decrease screen brightness
Left index + thumb pinch	Zoom in or zoom out (adjust page size)

🔄 How It Works

The system captures video input from your webcam and processes it through a computer vision pipeline:

1. Video Capture: Acquires real-time video stream from webcam
2. Hand Detection: MediaPipe tracks 21 hand landmarks with sub-pixel accuracy
3. Gesture Recognition: Analyzes hand configurations to identify specific gestures
4. Action Execution: Translates recognized gestures into corresponding mouse actions

👥 Use Cases

• Accessibility assistance for individuals with limited mobility
• Hands-free computing in sterile environments (medical/lab settings)
• Convenient multitasking during presentations or cooking
• Touchless public kiosk interfaces

📋 Requirements

Hardware:

• Webcam (720p, 30 FPS minimum)
• Intel i5/AMD Ryzen 5 or equivalent
• 4GB RAM (8GB recommended)

Software:

• opencv-python
• mediapipe
• numpy
• pyautogui

Troubleshooting

• Poor detection: Ensure proper lighting and clear hand visibility
• Cursor jitter: Increase smoothing factor in configuration
• Performance lag: Close resource-intensive applications or reduce frame rate

Built for accessible, intuitive human-computer interaction ❤️