Skip to content

Latest commit

 

History

History
41 lines (25 loc) · 2.01 KB

README.md

File metadata and controls

41 lines (25 loc) · 2.01 KB
Raw

Image Caption Generator with Streamlit, Hugging Face, and PyTorch

This repository contains a simple yet powerful image captioning application built with Streamlit, Hugging Face's transformers, and PyTorch. The application generates a descriptive caption for any uploaded image.

Overview

The Image Caption Generator is a demonstration of integrating the Salesforce/blip-image-captioning-large model, a pretrained transformer model from Hugging Face, the PyTorch library, and Streamlit for creating interactive web applications. The application generates captions that provide a textual representation of the images, leading to a more comprehensive and engaging user experience. Importantly, this feature-rich image captioning application is implemented in less than 50 lines of code (excluding whitespace and comments)!

Key Features

  • Streamlit: A powerful, fast Python framework used to create the web interface for the application.
  • Hugging Face's Salesforce/blip-image-captioning-large: A state-of-the-art image captioning AI model that generates the captions for the images.
  • PyTorch: A Python library used to load and interact with the pretrained AI model.

How to Run

Prerequisites

  • Python 3.6 or higher
  • Streamlit
  • PyTorch
  • HuggingFace Transformers

Steps

  1. Clone this repository: git clone https://github.com/AustonianAI/image-caption-generator.git
  2. Navigate to the project directory: cd image-caption-generator
  3. Install the necessary Python packages using the command pip install -r requirements.txt.
  4. Run the Streamlit app using the command streamlit run app.py.

Usage

The Image Caption Generator allows users to upload an image. After the image has been uploaded, it generates a descriptive caption for the image based on the content and context of the image.

Contribution

Contributions, issues, and feature requests are welcome. Feel free to check the Issues page if you want to contribute.

License

This project is licensed under the terms of the MIT license.