This project implements a computer vision system that detects faces in images and classifies them into three categories: with_mask, without_mask, and mask_incorrect. Using transfer learning with MobileNetV2, the model achieves real-time face mask detection capabilities.
🚀 Key Features
Object Detection: Extracts face regions from images using bounding box annotations
Multi-class Classification: Classifies faces into three mask-wearing categories
Deep Learning: Utilizes MobileNetV2 with custom classification head
Data Augmentation: Implements image transformations to improve model robustness
Performance Evaluation: Comprehensive metrics including confusion matrix and classification reports
