🧠 AuraScan

This repository contains an AI-powered AR Health Scanner and Virtual Try-On System.
It combines computer vision, deep learning, and augmented reality to detect human health cues, recognize objects, and overlay virtual items in real-time.

---

📂 Project Structure

├── model/ # Pre-trained & custom-trained ML/DL models ├── snaps/ # Example screenshots & output snapshots ├── tryon resources/ # Glasses / accessories / virtual objects for AR TryOn ├── error_handling.py # Centralized error handling & logging ├── health_scanner.py # AR Health Scanner (face, posture, nails, skin) ├── obj_recog.py # Object recognition using YOLO / vision models ├── skin_scan.py # Skin condition prediction (CNN) ├── train_labels.py # Dataset training & label mapping ├── tryon.py # AR Virtual Try-On engine └── README.md # Project documentation

---

🚀 Features

• Real-time AR Health Scanner

  • Eye fatigue detection
  • Posture analysis (spine, shoulders, head tilt)
  • Nail & skin health checks

• Skin Condition Classification

  • Uses a CNN trained on medical skin datasets
  • Detects and classifies conditions with accuracy reports

• Object Recognition

  • YOLOv8 / Mediapipe-based real-time recognition
  • Identifies daily objects, living/non-living items

• Virtual Try-On (AR)

  • Overlay glasses / accessories on user’s face
  • Uses tryon resources/ assets

• Robust Error Handling

  • Centralized logging in error_handling.py

• Training Utilities

  • train_labels.py handles dataset preprocessing & label mappings

---

🛠️ Installation

Requirements

• Python 3.9+
• OpenCV
• Mediapipe
• TensorFlow / Keras
• NumPy
• Ultralytics YOLO (optional for obj_recog.py)

Install dependencies

pip install -r requirements.txt