METU Artificial Intelligence Society – Face Recognition Project

The project focuses on building a face analysis system in three main stages using deep learning and computer vision tools like MediaPipe and OpenCV.

Project Roadmap

• Sprint 1 (Completed): Face detection, alignment, and cropping
• Sprint 2 (Planned): Face recognition and identity matching
• Sprint 3 (Planned): Facial attribute detection (emotion, age, gender, etc.)

Features (Sprint 1)

• Real-time face detection using MediaPipe
• Alignment of rotated faces
• Cropped and consistently scaled face output
• Interactive OpenCV trackbars for adjusting face scale and vertical shift

Requirements

• Python 3.6+
• OpenCV
• MediaPipe
• NumPy

Installation

git clone https://github.com/b454k/face-recognition.git
cd face-recognition
pip install opencv-python mediapipe numpy
python main.py