FridgePal 🧊

An AI-powered Smart Fridge Assistant that helps you manage your fridge inventory, discover recipes, and track nutrition—all with offline AI capabilities.

FridgePal combines computer vision, natural language processing, and mobile-first design to transform how you manage food and plan meals.

Demo 🎥

Check out the demo of FridgePal in action: Watch Demo

Features ✨

🎯 Core Features

• Smart Item Detection - Automatically detect and catalog items by taking photos of your fridge
• Inventory Management - Track food items, quantities, and expiration dates
• Smart Recipe Discovery - Get personalized recipe recommendations based on what you have
• Nutrition Tracking - Monitor daily calorie and macro intake
• Expiry Alerts - Get notified before items expire
• Food Recognition - Powered by YOLOv8 object detection and Google Gemini AI

📊 User Capabilities

• Profile Management - Set dietary goals, preferences, and health metrics
• Calorie & Macro Tracking - Monitor daily nutritional intake with detailed breakdowns
• Meal Planning - Plan meals and track what you've prepared
• Recipe Customization - Save favorite recipes and view personalized recommendations
• Water Intake Tracking - Stay hydrated with daily water reminders

Tech Stack 🛠️

Frontend

• React Native (v0.83.1) - Cross-platform mobile development
• Redux Toolkit - State management
• React Navigation - Navigation and routing
• Supabase Client - Backend authentication and database
• Google Gemini AI - Recipe generation and food insights

Backend

• FastAPI - Modern Python web framework
• YOLOv8 (Ultralytics) - Object detection for food items
• Tesseract OCR - Optical character recognition for nutrition labels
• PyTorch - Deep learning inference
• OpenCV - Image processing

Database

• Supabase (PostgreSQL) - User auth, profiles, and data storage

Platforms

• iOS - Native iOS app with Swift integration
• Android - Native Android with React Native bridge

Project Structure 📁

FridgePal/
├── frontend/
│   ├── src/
│   │   ├── components/       # Reusable UI components
│   │   ├── pages/            # Screen components (Auth, Home, Scanner, etc.)
│   │   ├── services/         # API and service integrations
│   │   ├── hooks/            # Custom React hooks
│   │   ├── config/           # Configuration (backend, supabase)
│   │   ├── store/            # Redux store and slices
│   │   └── utils/            # Helper functions
│   ├── Navigation/           # Navigation structure
│   ├── package.json
│   └── tsconfig.json
├── backend/
│   ├── main.py               # FastAPI application
│   ├── services/
│   │   ├── detection.py      # YOLOv8 food detection
│   │   ├── ocr.py            # Tesseract OCR
│   │   ├── food_inference.py # Food classification
│   │   └── __init__.py
│   ├── models/               # Pydantic schemas
│   ├── utils/                # Image processing utilities
│   ├── requirements.txt
│   └── config.py
├── ios/                      # iOS app configuration
├── android/                  # Android app configuration
├── supabase/
│   └── schema.sql            # Database schema
└── README.md

Getting Started 🚀

Prerequisites

• Node.js (v16+) and npm
• Python (3.9+) with pip
• React Native CLI
• Xcode (for iOS development)
• Android Studio (for Android development)
• Supabase Account (or local setup)

Installation

1. Clone the Repository

git clone https://github.com/Pallavi-Madhu/FridgePal.git
cd FridgePal

2. Set Up Backend

cd backend

# Create virtual environment
python -m venv venv

# Activate virtual environment
# On Windows:
venv\Scripts\activate
# On macOS/Linux:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Create .env file with your configuration
cp .env.example .env  # If available, or create manually

Required Environment Variables:

SUPABASE_URL=your_supabase_url
SUPABASE_KEY=your_supabase_key
GOOGLE_GENAI_KEY=your_gemini_api_key

3. Set Up Frontend

# Install Node dependencies
npm install

# Create .env or use react-native-dotenv
# Configure your backend URL and Supabase credentials

4. Set Up Database

1. Create a Supabase project at supabase.com
2. Run the SQL schema:

# Copy contents of supabase/schema.sql into Supabase SQL Editor
# Or use Supabase CLI
supabase db push

Running the Application 🎬

Backend Server

cd backend
python -m uvicorn main:app --reload --port 8000

The API will be available at http://localhost:8000

API Documentation: http://localhost:8000/docs

Frontend (Development)

Start Development Server

npm start

Run on Android

npm run android

Run on iOS

npm run ios

API Endpoints 📡

Detection Service

• POST /detect - Detect food items in image
  • Input: Base64 encoded image
  • Output: Detected items with bounding boxes and confidence scores

OCR Service

• POST /ocr - Extract text from nutrition labels
  • Input: Image region
  • Output: Recognized text

Food Inference

• POST /food-classify - Classify and get nutritional info
  • Input: Detected food items
  • Output: Food names with nutritional data

Gemini Integration

• POST /generate-recipes - Get recipe recommendations
  • Input: Available ingredients
  • Output: Recipe suggestions with instructions

Key Components 🧩

Frontend Pages

• Auth - Sign in and registration
• Home - Dashboard with fridge overview
• Scanner - Camera interface for item detection
• Recipe - Recipe discovery and saved recipes
• CalTracker - Nutrition and calorie tracking
• Profile - User settings and preferences

Redux Slices

• userSlice - User authentication and profile
• fridgeSlice - Fridge inventory management
• recipeSlice - Recipe management
• nutritionSlice - Nutrition tracking

Backend Services

• DetectionService - YOLOv8 model inference
• OCRService - Tesseract OCR wrapper
• FoodInferenceEngine - Food classification and nutrition lookup
• ImageUtils - Image cropping and preprocessing

Database Schema 🗄️

Main Tables

• profiles - User profiles with health goals
• fridge_items - Tracked food items with expiry dates
• recipes - Recipe database with ingredients and instructions
• saved_recipes - User's saved recipes (junction table)
• user_meals - Tracked meals with dates
• nutrition_logs - Daily nutrition data

See supabase/schema.sql for complete schema.

Development 💻

Running Tests

npm test              # Frontend tests
cd backend && pytest  # Backend tests

Linting

npm run lint

Code Structure Guidelines

• Use TypeScript in React components where possible
• Follow Redux pattern for state management
• Use custom hooks for reusable logic
• Implement error handling in API calls
• Add PropTypes or TypeScript types for components

Key Technologies Deep Dive 🔬

Computer Vision Pipeline

1. Image Capture - React Native Vision Camera
2. Preprocessing - OpenCV for normalization
3. Detection - YOLOv8 for food identification
4. OCR - Tesseract for nutrition label reading
5. Classification - Fine-tuned model for food names

AI Features

• Food Recognition - YOLOv8 trained on food datasets
• Recipe Generation - Google Gemini for contextual recipes
• Nutrition Data - Database lookup + AI enhancement
• Personalization - Redux-based user preferences

Configuration 🔧

Backend Config (backend/config.py)

• Model paths
• API endpoints
• Database connections
• CORS settings

Frontend Config (src/config/)

• backend.js - Backend API URL
• supabase.js - Supabase credentials and settings

Deployment 📦

Backend Deployment

• Deploy to Heroku, AWS Lambda, or Docker
• Set environment variables for production
• Configure database URLs

Frontend Deployment

• Build APK for Android: ./gradlew build
• Build IPA for iOS using Xcode
• Distribute via app stores or TestFlight

Contributing 🤝

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Known Limitations & Future Work 🚧

Current Limitations

• YOLOv8 detection performance depends on lighting conditions
• OCR accuracy varies with label quality
• Offline mode requires pre-downloaded models

Planned Features

• 🌐 Offline mode with cached recipes
• 🤖 Machine learning for personalized recommendations
• 📱 Widget support for quick inventory checks
• 🌍 Multi-language support
• 🍽️ Integration with meal planning services
• 📊 Advanced nutrition analytics

Troubleshooting 🔧

Backend Issues

• Model loading errors: Ensure YOLOv8 model is downloaded
• CORS errors: Check frontend URL in CORS middleware
• Database connection: Verify Supabase credentials in .env

Frontend Issues

• Camera permissions: Ensure app has camera access granted
• Image processing slow: Reduce image resolution
• Supabase auth errors: Check API keys and authentication config

License 📄

This project is licensed under the MIT License - see LICENSE file for details.

Support & Contact 💬

For issues, questions, or suggestions:

• Open an issue on GitHub
• Contact the development team

Acknowledgments 🙏

• YOLOv8 by Ultralytics
• Google Gemini AI
• Supabase for backend services
• React Native community

---

Made with ❤️ by the FridgePal Team