PicPeek

PicPeek is an intelligent image analysis companion that harnesses the power of GPT-4 vision to decode and understand images. This Streamlit-based web application offers a user-friendly interface for exploring AI-powered visual understanding.

Features

• Image Caption Generation: Get concise captions for your images.
• Detailed Description: Receive thorough, detailed descriptions of image content.
• Image Classification: Categorize images into predefined categories.
• Object Recognition: Identify multiple objects within an image.
• Subject Identification: Determine the primary subject of an image.
• Creative Writing: Generate short stories based on image content.
• Emotion Analysis: Interpret emotions displayed by people in images.
• Text Transcription: Extract and transcribe text found in images.
• Text Translation: Translate text in images to different languages.

Installation

1. Clone the repository:

   git clone https://github.com/bdeva1975/picpeek.git
   cd picpeek
   

2. Install the required dependencies:

   pip install -r requirements.txt
   

3. Set up your OpenAI API key as an environment variable:

   export OPENAI_API_KEY='your-api-key-here'
   

Usage

1. Run the Streamlit app:

   streamlit run image_understanding_app.py
   

2. Open your web browser and navigate to the provided local URL (typically http://localhost:8501).

3. Select an image from the provided examples or upload your own.

4. Choose a prompt type or enter a custom prompt.

5. Click "Go" to generate the analysis.

Project Structure

• image_understanding_app.py: Main Streamlit application file.
• image_understanding_lib.py: Library containing helper functions for image processing and API interactions.
• images/: Directory containing example images.
• requirements.txt: List of Python dependencies.

Dependencies

• streamlit
• openai
• Pillow

Contributing

Contributions to PicPeek are welcome! Please feel free to submit a Pull Request.

License

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

Acknowledgements

• This project uses the OpenAI GPT-4 vision model for image analysis.
• Streamlit for the web application framework.

Disclaimer

This application is for educational and demonstration purposes only. Please ensure you comply with OpenAI's use-case policies when using this application.