Claude Vision Object Detection

A powerful Python tool that leverages Claude 3.5 Sonnet Vision API to detect and visualize objects in images. The script automatically draws bounding boxes around detected objects, labels them, and displays confidence scores.

Features

🖼️ Process single images or entire directories
📦 Automatic object detection with bounding boxes
🎯 High-precision confidence scores
🎨 Vibrant, distinct colors for each detected object
💾 Saves annotated images with detection results

Requirements

Python 3.7+
Anthropic API key
Required Python packages (see requirements.txt)

Installation

Clone the repository:

git clone https://github.com/doriandarko/claude-vision-object-detection.git
cd claude-vision-detection

Install required packages:

pip install -r requirements.txt

Create a .env file in the project root and add your Anthropic API key:

ANTHROPIC_API_KEY=your_api_key_here

Usage

Run the script:

python main.py

When prompted, enter either:
- Path to a single image file
- Path to a directory containing multiple images
The script will:
- Process each image using Claude Vision API
- Draw bounding boxes around detected objects
- Add labels with confidence scores
- Save annotated images in an output directory

Supported Image Formats

JPEG (.jpg, .jpeg)
PNG (.png)
GIF (.gif)
WebP (.webp)

Output

The script creates an output directory in the current working directory. Processed images are saved with the prefix detected_ followed by the original filename.

Error Handling

The script includes comprehensive error handling for:

Invalid image paths
Unsupported file formats
API communication issues
Image processing errors

Contributing

Contributions are welcome! Please feel free to submit pull requests or create issues for bugs and feature requests.

License

This project is licensed under a modified MIT License with attribution requirements. See the LICENSE file for details.

Attribution Requirements

When using this software or its derivatives, you must include:

The name of the original project (Claude Vision Object Detection)
A link to the original repository
The name of the original author

Acknowledgments

Built using the Claude 3.5 Sonnet Vision API by Anthropic
Uses PIL (Python Imaging Library) for image processing
Implements the golden ratio for color generation

Support

For issues, questions, or contributions, please:

Check existing GitHub issues
Create a new issue with a detailed description
Include sample images if relevant (without sensitive data)

Note: This tool relies on the Claude Vision API, which requires an API key from Anthropic. Make sure you have appropriate access and credits before using.

Doriandarko/Claude-Vision-Object-Detection