Reverse Image Dataset Generator 🖼️ → 📝

Create high-quality training datasets for image generation models by automatically generating optimized prompts from your images using advanced GPT models.

✨ Features

• 🤖 AI-Powered Analysis: Automatically generate optimized prompts using GPT models
• 🎯 Context-Aware: Include custom context to guide prompt generation
• 📦 Batch Processing: Handle multiple images simultaneously
• 🏷️ Smart Tagging: Auto-generate relevant tags for better organization
• 📤 Standard Format: Export datasets in JSONL format ready for fine-tuning
• 👀 Live Preview: Real-time visualization of generated prompts and tags
• 🔄 Progress Tracking: Monitor batch processing with detailed progress updates

🚀 Quick Start

Prerequisites

• Node.js (v18 or later)
• NPM (v8 or later)
• OpenAI API key
• PostgreSQL database (setup guide)

Installation

1. Clone and install dependencies:

git clone https://github.com/yourusername/reverse-image-dataset-generator.git
cd reverse-image-dataset-generator
npm install

2. Set up your environment:

cp .env.example .env

3. Add your OpenAI API key to .env:

OPENAI_API_KEY=your_api_key_here

4. Start the development server:

npm run dev

Visit http://localhost:5050 to start using the application.

💡 Use Cases

Fine-Tuning Image Generation Models

• Create custom training datasets for models like OmniGen
• Generate high-quality prompts from existing image collections
• Maintain consistency in prompt style across datasets

Dataset Management

• Convert image libraries into structured training data
• Add context-aware descriptions to image collections
• Generate semantic tags for better organization

Content Creation

• Generate optimized prompts for existing images
• Create consistent image-prompt pairs
• Build reference libraries for prompt engineering

📖 How It Works

1. Upload Images

   • Drag & drop or select multiple images
   • Add optional context for better descriptions
   • Preview selected images instantly

2. AI Processing

   • Images analyzed by GPT models
   • Context-aware prompt generation
   • Automatic tag extraction
   • Real-time progress tracking

3. Dataset Creation

   • JSONL file generation with prompts
   • Original images included
   • ZIP archive for easy download
   • Standard format for fine-tuning

Example output format:

{
  "task_type": "text_to_image",
  "instruction": "a serene landscape with snow-capped mountains reflecting in a crystal-clear lake at sunset",
  "input_images": [],
  "output_image": "mountain_lake.jpg"
}

🛠️ Technical Details

Architecture

• Frontend: React + TypeScript + Shadcn UI
• Backend: Express.js + Multer
• Database: PostgreSQL (configuration guide)
• AI: OpenAI GPT-4o series models
• Storage: File system with organized structure

Available Models

• gpt-4o-mini: Fast, efficient processing
• gpt-4o: Balanced performance
• gpt-4o-2024-11-20: Latest model with enhanced capabilities

🤝 Contributing

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

📝 License

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

🌟 Next Steps

• Add user authentication
• Implement dataset management UI
• Add advanced image processing options
• Optimize batch processing
• Add dataset search and filtering

🔗 Links

• Documentation
• Database Guide
• API Reference
• Contributing Guidelines

Docker Setup

Port Configuration

The application uses the following default ports:

• 5050: Main application server
• 5173: Vite development server (development only)
• 5432: PostgreSQL database

To avoid port conflicts, you can customize these ports in your environment files:

PORT=5050            # Main application port
VITE_PORT=5173      # Vite dev server (development only)
DB_PORT=5432        # PostgreSQL port

Development Environment

1. Copy environment files:

   cp .env.development.example .env.development

2. Adjust ports if needed in .env.development

3. Start development environment:

   docker compose -f docker-compose.dev.yml --env-file .env.development up --build

The development environment includes:

• Hot reloading for both frontend and backend
• PostgreSQL database with automatic schema initialization
• Vite dev server for frontend development

Production Environment

1. Copy environment files:

   cp .env.production.example .env.production

2. Update production environment variables:

   • Set secure PostgreSQL password
   • Configure OpenAI API key
   • Adjust ports if needed
   • Adjust other settings as needed

3. Start production environment:

   docker compose -f docker-compose.prod.yml --env-file .env.production up -d --build

The production environment includes:

• Optimized production build
• PostgreSQL database with persistent storage
• Automatic container restart
• Health checks for all services

Database Management

The database is automatically initialized with the schema and migrations when the container starts. To manage the database:

1. Generate new migrations:

   docker compose -f docker-compose.dev.yml --env-file .env.development exec app npm run db:generate

2. Apply migrations:

   docker compose -f docker-compose.dev.yml --env-file .env.development exec app npm run db:push

3. Access database:

   docker compose -f docker-compose.dev.yml --env-file .env.development exec db psql -U postgres

Environment Variables

See .env.development.example and .env.production.example for required environment variables.

Volumes

• Development: pgdata_dev - PostgreSQL data (development)
• Production: pgdata_prod - PostgreSQL data (production)

Ports

• Server: 5050
• Vite Dev Server: 5173 (development only)
• PostgreSQL: 5432