https://github.com/mjahmadee/image_captioning

Image Captioning
https://github.com/mjahmadee/image_captioning

computer-vision image-captioning nlp

Last synced: 26 days ago
JSON representation

Image Captioning

• Host: GitHub
• URL: https://github.com/mjahmadee/image_captioning
• Owner: MJAHMADEE
• License: mit
• Created: 2023-07-15T05:21:39.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2024-03-16T12:46:13.000Z (over 2 years ago)
• Last Synced: 2025-02-28T20:45:39.121Z (over 1 year ago)
• Topics: computer-vision, image-captioning, nlp
• Language: Jupyter Notebook
• Homepage:
• Size: 22.6 MB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Image Captioning with Neural Networks 🖼️🤖

![Python](https://img.shields.io/badge/Python-3.8-blue.svg)

![PyTorch](https://img.shields.io/badge/PyTorch-1.8.1-orange.svg)

![License](https://img.shields.io/badge/License-MIT-green.svg)

Image Captioning with Neural Networks is a deep learning project that combines Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) to generate captions for images automatically. This implementation utilizes a pre-trained ResNet model for image feature extraction and an LSTM network for generating textual descriptions of the images.

## Features 🌟

- Utilizes a pre-trained ResNet-18 model for efficient image feature extraction.

- Employs an LSTM network for generating descriptive captions based on image features.

- Supports training with and without fine-tuning of the ResNet model.

- Includes functionality for both training and testing the model with a custom dataset.

- Visualizes training loss and sample predictions to assess model performance.

## Setup and Installation 🛠️

1. Clone the repository from GitHub.

2. Navigate to the project directory.

3. Install the required dependencies listed in the `requirements.txt` file.

## Dataset 📁

The model is trained and tested on the Flickr8k dataset, which comprises 8,000 images each paired with five different captions. For the purpose of this project, the dataset is pre-processed to align with the model's requirements.

## Training the Model 🚀

Training the model involves executing the training script, which will start the training process and save the model weights periodically.

## Testing the Model 🧪

After training, the model's performance can be evaluated by executing the testing script, which generates captions for the images in the test dataset.

## Results and Evaluation 📊

The model's performance can be evaluated based on the captions generated for the test images. A qualitative assessment involves comparing the predicted captions against the ground truth captions.

## License 📜

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

## Acknowledgements 🙌

- Thanks to the creators of the Flickr8k dataset for providing the resources necessary for training and testing the model.

- PyTorch documentation for providing comprehensive guides and tutorials.

## Notebook and Copyright



@misc{MJImageCaptioning2023,

  author = {Mohammad Javad (MJ) Ahmadi},

  title = {Image Captioning},

  year = {2023},

  publisher = {GitHub},

  journal = {GitHub repository},

  howpublished = {\url{https://github.com/MJAHMADEE/Image_Captioning}}

}

---

For more information, please refer to the [official repository](https://github.com/MJAHMADEE/Image_Captioning).