This project uses AI to detect your face shape through a webcam and recommends personalized hairstyles and glasses based on your facial features and structure.

[Coming Soon]

• Frontend: React, React Router, OpenCV.js, Webcam
• Backend: Flask, PyTorch, VGG16
• Model: Pre-trained VGG16 fine-tuned for face shape classification

• Face detection with webcam
• Face shape classification using a fine-tuned VGG16 model
• Personalized hairstyle and glasses suggestions for both male and female users
• Interactive UI to view results and try different looks

Accepts a base64-encoded image and selected gender, returns a predicted face shape and corresponding hairstyle recommendations.

{
  "image": "data:image/jpeg;base64,...",
  "gender": "male" | "female"
}

{
  "result": "oval",
  "hairstyles": [
    {
      "id": 21,
      "name": "Buzz Cut",
      "image": "/images/male/buzz-cut.jpg"
    },
    ...
  ],
  "glasses": [
    {
      "id": 29,
      "name": "Round",
      "image": "/glasses/Round.png"
    },
    ...
  ]
}

• Webcam Capture: Users capture a selfie directly through their webcam.
• Face Detection: The app uses OpenCV.js for detecting the face area.
• Face Shape Prediction: The backend runs the captured face through a fine-tuned VGG16 model to classify the user's face shape (oval, heart, square, round, oblong).
• Personalized Recommendations: Based on the face shape and gender, the system provides:
  • A set of recommended hairstyles.
  • A set of recommended glasses.
• Interactive Results Page: Users can view and explore different recommended styles and eyewear.

Yahia Alqurnawi @Yahialqur

Jon Kruja @Spec1alKs