|
|
--- |
|
|
title: Face Shape Detection |
|
|
emoji: 👤 |
|
|
colorFrom: blue |
|
|
colorTo: purple |
|
|
sdk: docker |
|
|
pinned: false |
|
|
license: mit |
|
|
app_port: 7860 |
|
|
short_description: AI face shape detection with MediaPipe & ML |
|
|
--- |
|
|
|
|
|
# Face Shape Detection |
|
|
|
|
|
An AI-powered application that analyzes facial features to determine face shape using MediaPipe landmarks and machine learning. |
|
|
|
|
|
## Features |
|
|
|
|
|
- **Real-time face shape detection** from uploaded images |
|
|
- **5 face shape categories**: Heart, Oval, Round, Square, Oblong |
|
|
- **Facial measurements** with confidence scores |
|
|
- **Interactive web interface** with image upload |
|
|
- **RESTful API** for integration |
|
|
|
|
|
## How it Works |
|
|
|
|
|
1. **Face Detection**: Uses MediaPipe to detect and extract facial landmarks |
|
|
2. **Feature Extraction**: Calculates key facial measurements and ratios |
|
|
3. **ML Classification**: Uses a trained Random Forest model to predict face shape |
|
|
4. **Results**: Returns face shape, confidence scores, and facial measurements |
|
|
|
|
|
## API Endpoints |
|
|
|
|
|
- `POST /analyze` - Upload an image for face shape analysis |
|
|
- `GET /` - Web interface for image upload |
|
|
- `GET /video_feed` - Real-time video feed (if camera available) |
|
|
|
|
|
## Usage |
|
|
|
|
|
1. Upload an image using the web interface |
|
|
2. The system will analyze the face and return: |
|
|
- Detected face shape |
|
|
- Confidence scores for all categories |
|
|
- Facial measurements (length, width, etc.) |
|
|
- Annotated image with landmarks |
|
|
|
|
|
## Technical Details |
|
|
|
|
|
- **Framework**: Flask |
|
|
- **Computer Vision**: MediaPipe, OpenCV |
|
|
- **ML Model**: Random Forest (scikit-learn) |
|
|
- **Image Processing**: Smart preprocessing with face detection |
|
|
- **Deployment**: Docker container optimized for Hugging Face Spaces |
|
|
|
|
|
## Model Performance |
|
|
|
|
|
The model uses optimized features extracted from 468 facial landmarks and achieves high accuracy in face shape classification across diverse face types. |
|
|
|
|
|
## Requirements |
|
|
|
|
|
- Python 3.10+ |
|
|
- MediaPipe |
|
|
- OpenCV |
|
|
- scikit-learn |
|
|
- Flask |
|
|
- NumPy |
|
|
|
|
|
## License |
|
|
|
|
|
MIT License - feel free to use and modify for your projects. |
