README.md

metadata

title: SmartVision AI
emoji: 👀
colorFrom: blue
colorTo: green
sdk: docker
pinned: false
license: mit
app_port: 7860

🔍 SmartVision: Advanced Object Detection & Classification

This project is an end-to-end Computer Vision pipeline designed for training and comparing multiple Deep Learning architectures (VGG16, ResNet50, MobileNetV2, EfficientNet) and real-time detection via YOLOv8.

🚀 Project Features

• Multi-Model Comparison: Dynamic leaderboard comparing accuracy and inference latency.
• YOLOv8 Integration: Real-time object detection on 25 COCO-subset classes.
• Automated Pipeline: Scripts for dataset generation, model training, and evaluation.

📂 Directory Structure

• app.py: Main Streamlit interface.
• models/: Contains the trained .h5 and .pt weights.
• pages/: Additional dashboard pages for performance analytics.
• training/: Specialized scripts for each architecture.

🛠️ Local Setup

1. Clone the repository.
2. Install dependencies: pip install -r requirements.txt
3. Run the app: streamlit run app.py