FitFindr is an AI-powered fashion recommendation app that personalizes outfit ideas based on your body shape, style, and confidence. Upload a photo or describe your vibe — FitFindr uses computer vision, Gemini AI, and web intelligence to curate outfits that fit you, not just your size.
👕 Smart Outfit Classification: YOLO model classifies clothing types from Pinterest images.
🧠 AI Style Reasoning: Gemini Pro explains why specific fits work for your body and aesthetic.
❤️ Personalized Recommendations: Combines your style input + AI reasoning to suggest curated outfits.
🔁 Feedback Loop: Like/dislike to refine your future recommendations.
💾 Lightweight Architecture: JSON datastore for simple, fast MVP performance.
Frontend: ReactTS + CSS + HTML Backend: Python (FastAPI) AI Models: YOLOv8, Google Gemini 2.5 Pro, CLIP Embeddings Data: Pinterest-scraped outfit images, hierarchical JSON fashion taxonomy
1️⃣ Clone the repo git clone https://github.com/yourusername/fitfindr.git cd fitfindr
2️⃣ Backend Setup cd backend pip install -r requirements.txt uvicorn main:app --reload
3️⃣ Frontend Setup cd frontend npm install npm start
4️⃣ Open in browser
Go to http://localhost:3000 and start exploring your AI stylist 🎨
We built FitFindr because fashion shouldn’t be intimidating or exclusive. People deserve clothes that celebrate who they are — not who the internet thinks they should be. So we created a bridge between AI and self-confidence — helping people discover their best fits effortlessly.