A real-time smile detection web application built with OpenCV, MediaPipe, and FastAPI. The app uses a webcam to detect faces and determine whether the user is smiling, displaying the results on a frontend built with vanila HTML/CSS.

• Real-time face and smile detection
• Accurate results using MediaPipe's face mesh
• Lightweight and fast performance via FastAPI backend
• Simple and clean web interface using HTML & CSS
• Project also includes files for detecting Faces and Hands.

• Python
  • opencv-python (cv2) for video capture and image processing
  • mediapipe for face mesh and smile detection logic, also tracks and detects lips.
  • fastapi for serving the backend and APIs
• Frontend
  • HTML and CSS for UI layout and styling

git clone https://github.com/Yeeyash/Smile-Detector-using-OpenCV.git

Start the FastAPI server:

uvicorn smilepage:app --reload

Then open your browser and navigate to:

http://127.0.0.1:8000

1. OpenCV captures real-time video from the webcam.
2. MediaPipe's face mesh model detects key facial landmarks. 3.A smile is detected by analyzing the distances and movement between key mouth landmarks.
3. Results are streamed to the browser using FastAPI endpoints.
4. The HTML/CSS frontend renders the webcam feed and updates smile status dynamically.

• OpenCV

• MediaPipe

• FastAPI

This project is licensed under the MIT License.