Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to content

oCaption: Leveraging OpenAI's GPT-4 Vision for Advanced Image Captioning

License

Notifications You must be signed in to change notification settings

ghostofpokemon/oCaption

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 

Repository files navigation

Image Captioning Tool with OpenAI GPT-4

This Python tool is designed to generate captions for a set of images, utilizing the advanced capabilities of OpenAI's GPT-4 Vision API. It can handle image collections either from a ZIP file or a directory. The tool offers flexibility in captioning, providing options to describe images directly or in a creative style, like "in the style of Family Guy."

Key Features

  • Processes images from ZIP files or directories.
  • Utilizes OpenAI's GPT-4 Vision API for caption generation.
  • Supports custom caption formats, including style-based descriptions (e.g., "in the style of TOK").
  • Captions are conveniently saved in a CSV file for easy access and reference.

Example Image

System Requirements

  • Python 3
  • An active OpenAI API key with GPT-4 Vision API access.
  • Necessary Python libraries: requests, imgcat.

Setup Instructions

  1. Clone the GitHub repository:

    git clone https://github.com/ghostofpokemon/oCaption.git
    
  2. Change to the tool's directory:

    cd oCaption
    
  3. Install the required Python dependencies:

    pip install requests imgcat
    

How to Use

Set your OpenAI API key as an environment variable:

export OPENAI_API_KEY="Your-API-Key-Here"

Launch the tool with Python:

python3 oCaption.py

Follow the prompts to input the path to your image folder or ZIP file, specify the TOK value (such as "TOK" or "Family Guy"), and choose a caption prefix if needed.

Output Details

The tool generates detailed captions for each processed image, saving them in a caption.csv file located in the current working directory. The captions follow the format "a photo of [subject]" or "in the style of [TOK]" based on your preference.

Contributing to the Project

We welcome contributions! To propose changes, please use the standard GitHub pull request process.

License Information

This tool is distributed under the MIT License.

Support and Contact

For queries or feedback, feel free to open an issue in the GitHub repository.

Releases

No releases published

Languages