SmartChef is an intelligent web app that helps users discover recipes using computer vision and generative AI.
Simply take or upload a photo of your fridge or ingredients, and SmartChef will identify the food items using YOLOv8 and generate creative, personalized recipes using Google Gemini AI.
- 🧠 AI Object Detection — Detects ingredients from uploaded or live camera images using YOLOv8 Nano.
- 🤖 Recipe Generation — Generates custom recipes through Google Gemini API.
- 📸 Live Camera Mode — Take pictures directly from your webcam or phone camera.
- 🌐 Web Integration — FastAPI backend connected to a responsive web frontend hosted on Render and GitHub Pages.
- 🧾 Fallback Recommendations — Simple CSV-based recipe database for offline recommendations.
| Layer | Technology |
|---|---|
| Frontend | HTML, CSS, JavaScript (Fetch API for backend calls) |
| Backend API | FastAPI (Python) |
| AI / ML | YOLOv8 (Ultralytics), Google Gemini |
| Deployment | Render (Backend), GitHub Pages (.tech frontend) |
| Data | CSV-based sample recipe storage |
Hackathon/ │ ├── smartchef/ │ ├── api/ │ │ ├── gemini_generator.py # Handles Gemini recipe generation │ │ └── test_key.py # API key test file │ │ │ ├── edge/ │ │ └── detector.py # YOLOv8 food detection logic │ │ │ ├── ui/ │ │ └── app.py # FastAPI app (main backend entrypoint) │ │ │ └── cloud/ │ └── sample_recipes.csv # Local fallback recipe database │ ├── index.html # Web interface ├── script.js # Handles camera + API requests ├── style.css # Styling ├── requirements.txt # Python dependencies └── README.md # You are here 🙂
Tyler Lee - AI & Backend Integration, Full stack Ariful Shayun - Frontend, Testing Bryce Wong - Frontend & Web Design Saffat Uddin - Frontend & Web Design