eyeris

A high-performance image analysis service that leverages AI models to provide detailed analysis of images. Supports multiple AI providers (currently OpenAI and Ollama).

Features

• 🚀 High-performance image processing with parallel execution
• 🤖 Multiple AI provider support (OpenAI, Ollama)
• 📊 Detailed image analysis with customizable formats
• 🖼️ Automatic image optimization and thumbnail generation
• 📈 Token usage tracking
• 🔒 Rate limiting built-in
• 📝 Comprehensive logging

Quick Start

# Clone the repository
git clone https://github.com/scald/eyeris

# Navigate to project directory
cd eyeris

# Configure OpenAI API key (if using OpenAI)
echo "OPENAI_API_KEY=your_key_here" > .env

# Run the service
cargo run

API Usage

Send an image for analysis:

curl -X POST http://localhost:3000/process \
  -F "image=@path/to/your/image.jpg" \
  -F "provider=ollama" \
  -F "model=moondream" \
  -F "format=json"

Output Formats

• JSON: Structured data format for programmatic use
• Concise: Brief summary of key elements
• Detailed: Comprehensive analysis with multiple aspects
• List: Enumerated format for easy reading

Performance

The service is optimized for performance through:

• Parallel processing with Rayon
• Automatic image optimization
• Efficient memory management
• Rate limiting for stability

Project Structure

src/
├── processor.rs # Core processing logic
├── providers/   # AI provider implementations
├── prompts.rs   # Analysis prompt templates
└── main.rs      # Service entry point

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Add your changes
4. Write/update tests
5. Submit a pull request

License

MIT License

Built with Rust 🦀