Face Mesh with Mediapipe and OpenCV

This project demonstrates real-time 3D face landmark detection using MediaPipe FaceMesh and OpenCV. The program draws the full face mesh, highlights the nose tip, eyes, and mouth, and displays the FPS counter.

---

Features

• Capture live video from a webcam.
• Render the 468-point face mesh.
• Highlight the nose tip with a red circle.
• Highlight eyes in blue.
• Highlight mouth in orange.
• Show the FPS in the top-left corner.
• Mirror the video feed (natural mirror effect).
• Close the program with the Esc key.

---

Installation

1. Clone this repository or save the script locally.
2. Install the required packages:

pip install -r requirements.txt

Run

Launch the script with:

python SMP.py

Once started, the webcam window will open showing:

• The face mesh overlay
• The red nose tip marker
• Eyes in blue
• Mouth in orange
• The FPS counter

Press Esc to exit.

---

Useful Links

• MediaPipe FaceMesh Documentation
• OpenCV Documentation