🖱️ Virtual Mouse using Eye Tracking

A Python-based virtual mouse system that allows users to control their cursor and perform click actions using eye movement and blinking, powered by OpenCV, MediaPipe, and PyAutoGUI.

🚀 Features

• Cursor movement using eye gaze (tracked via the right eye).
• Left-click action triggered by blinking (detected via the left eye).
• Basic calibration support (placeholder logic included).

🧠 How It Works

• Uses MediaPipe's Face Mesh to detect facial landmarks.
• Tracks the right eye (landmark 474–478) for cursor control.
• Monitors left eye (landmarks 145 and 159) to detect blinking.
• If the eye remains closed beyond a threshold, a click is triggered.
• Smoothing is applied to prevent cursor jitter.

🧰 Tech Stack

• OpenCV – Real-time computer vision.
• MediaPipe – Facial landmark detection.
• PyAutoGUI – Cursor control and automation.

🎮 Controls

• Move Cursor: Look around using your right eye.
• Click: Blink your left eye (hold it shut briefly).
• Calibrate: Press c (Currently in development).
• Quit: Press q.

📌 Notes

• Works best in well with one who is not using spectacles.
• Ensure your webcam is unobstructed.
• Try adjusting the blink threshold if false positives occur.

👨‍💻 Author

Falak Rana
LinkedIn • GitHub