🖼️ Image Caption Generator with Gemini

A Streamlit web application that generates captions for uploaded images using Google's Gemini AI model.

Features

• Upload images in JPG, JPEG, or PNG format
• Choose from three description styles:
  • Short (1-line description)
  • Detailed (paragraph description)
  • Poetic (caption as a poem)
• Instant AI-powered image descriptions
• Simple and intuitive user interface

Prerequisites

Before running this application, you'll need:

• Python 3.7+
• A Google AI Studio API key (Gemini API)
• Obtain a Google AI Studio API key from here

Installation

1. Clone this repository or download the source code:

git clone https://github.com/DSC-UIT-khi/build-with-ai.git
cd build-with-ai

2. Create a virtual environment (recommended):

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install the required packages:

pip install -r requirements.txt

Configuration

Replace the API key in the code with your own Google AI Studio API key:

client = genai.Client(api_key="YOUR_API_KEY_HERE")

Important: It's recommended to use environment variables for API keys in production environments.

Usage

1. Run the Streamlit application:

streamlit run image_caption_app.py

2. Access the application in your web browser (typically at http://localhost:8501)

3. Upload an image using the file uploader

4. Select your preferred description style from the dropdown menu

5. Click "Generate Description" to get your AI-generated caption

How It Works

1. The application accepts image uploads from the user
2. The uploaded image is saved to a temporary file
3. The image is sent to Google's Gemini model along with a prompt based on the selected description style
4. Gemini analyzes the image and generates a description
5. The description is displayed on the web interface

Dependencies

• Streamlit - Web application framework
• Google Generative AI - AI model for image analysis and caption generation

Security Note

The current code includes an API key directly in the source. For production use:

1. Move the API key to an environment variable or a secure configuration file
2. Add any configuration files with sensitive information to .gitignore

Contribute

Contributions are welcome! Feel free to submit issues or pull requests.

Author

Made with ❤️ by Moosa Raza

License

This project is open source and available under the MIT License.

Acknowledgements

• Google for providing the Gemini AI model
• Streamlit for the excellent web app framework