Image From Text Generation 🎨

🎞️Demo

📝Content

• Overview
• Technologies
• Features
• Project Structure
• Setup and Run
• Contributing
• Contact

🗺️Overview

This project develops a text-to-image generation model that translates textual descriptions into corresponding images using deep learning techniques such as Convolutional Neural Networks (CNNs) and Transformers.

👨‍💻Technologies

• TensorFlow: Main framework for building the model.
• Python: Programming language used.
• Jupyter Notebook: For interactive development and testing.
• NumPy: For numerical operations.
• Matplotlib: For generating visualizations.
• Pandas: For data manipulation and analysis.
• Pillow: For image processing tasks.
• TQDM: For providing progress bars in loops.
• Requests: For making HTTP requests in Python.
• Scikit-Learn: For machine learning tools.
• Gradio: For building a web interface to interact with the model.
• COCO Dataset: Utilized for training and validating the model performance.

👀Features

• Dynamic Image Generation: Transforms text descriptions into high-quality images using CNNs and Transformers.
• Interactive Model Training: Detailed workflow in Jupyter Notebook using the COCO dataset for training and validating.

🗃️Project Structure

• image-generation-from-text-gui.ipynb: The main Jupyter notebook containing all the code for training the model and generating images using CNNs and Transformers.

🛠Setup and Run

1. Clone the repository:

git clone https://github.com/buroff1/image-from-text-generation.git
cd image-from-text-generation

2. Download the COCO dataset:

• Visit COCO Dataset and download the train2017 images and the annotations folder.
• Ensure these are placed within a data/ directory in the project root.

3. Launch Jupyter Notebook:

jupyter notebook

Open the image-generation-from-text-gui.ipynb notebook and follow the instructions within to train the model and generate images using CNNs and Transformers.

🤝Contributing

Interested in contributing to this project? Here's how you can help:

• Fork the repository.
• Create a new feature branch (git checkout -b feature/YourNewFeature).
• Make your changes and commit them (git commit -am 'Add some YourNewFeature').
• Push to the branch (git push origin feature/YourNewFeature).
• Submit a pull request.

📧Contact

For any inquiries or collaboration requests, please contact:

• Email: artem.burov0205@gmail.com
• GitHub: buroff1