Fitzy Virtual Try-On

See yourself wearing real, purchasable outfits before you buy.

Overview

Virtual Try-On is an AI-powered fashion platform that helps users visualize outfits on their own photos using real products from online stores. Users upload a full-body image, enter a style prompt, and receive personalized outfit recommendations with realistic AI-generated try-ons.

Features

• Full-body photo upload
• Style prompt input (e.g., Korean Streetwear, CEO Look)
• AI-powered outfit recommendations
• Virtual try-on generation
• Direct shopping links to products
• Personal wardrobe integration for better suggestions
• Save favorite outfits and looks

Tech Stack

• Frontend: React
• Backend: FastAPI
• Database: MongoDB, Cloudinary
• AI/ML: Gemini 2.5 Flash (Style Feedback), ResNet-50 & ONNX Runtime (Similarity Search), Fashn VTON & PyTorch (Virtual Try-On)

Documentation

• Backend API Docs: backend/API_DOCS.md
• Fashion Similarity Search Engine: ml/SIMILARITY_SEARCH.md
• FashnVTON Try-On Optimization Report: ml/OPTIMIZATION_REPORT.md

Contributors

• Akshita — AI/ML
• Utkarsh — Backend
• Yashika — Frontend