Face Attractiveness Detection

This project provides a Python script to analyze the attractiveness of individuals in a group photo using the reverse-engineered Attractiveness Test Android API. The script utilizes the YuNet model for face detection and performs image processing to evaluate and display attractiveness scores for each detected face.

Features

• Detects faces in a group image.
• Rotates and aligns faces for better accuracy.
• Uploads individual face images to the Attractiveness Test API for analysis.
• Draws attractiveness scores directly on the original image.

Disclaimer

I am not affiliated with attractivenesstest.com in any way. The use of their API in this project is for educational purposes only. I am not responsible for any issues or consequences that may arise from using their services.

Requirements

opencv-python
numpy
requests
tqdm

Installation

To get started, clone this repository and install the required Python packages.

Clone the Repository

git clone https://github.com/Andrei-Iosifescu123/face-attractiveness-analysis.git
cd face-attractiveness-analysis

Install Dependencies

pip install -r requirements.txt

Usage

main.py [-h] --input INPUT [--output OUTPUT]

Example Usage:

python3 main.py --input examples/example1.jpg

Known Issues

The server can't evaluate some faces and returns {"error":"Face detected, but error during age detection 'with_padding'"}.