Neural Style Transfer App

Overview

This repository contains a Streamlit web application that implements Neural Style Transfer (NST), a technique that allows you to combine the content of one image with the style of another image. The implementation is based on the work of Leon Gatys et al. as described in their paper "A Neural Algorithm of Artistic Style".

What is Neural Style Transfer?

Neural Style Transfer is a technique that uses convolutional neural networks to separate and recombine the content and style of arbitrary images. The algorithm works by:

1. Using a pre-trained CNN (VGG19) to extract content features from a content image
2. Extracting style features (via Gram matrices) from a style image
3. Iteratively optimizing a new image to match both the content features of the content image and the style features of the style image

The result is a new image that preserves the structural content of one image while adopting the artistic style of another.

Examples

Here are examples of what this app can do:

Content Image	Style Image	Result

Features

• User-friendly web interface built with Streamlit
• Support for uploading custom content and style images
• Option to choose between low-resolution (faster) and high-resolution (slower) processing
• Real-time preview of uploaded images
• Progress tracking during the style transfer process

Installation

Prerequisites

• Python 3.7 or later
• PyTorch
• CUDA-capable GPU (recommended for faster processing)

Clone the Repository

git clone https://github.com/amitch2019/NST_torch.git
cd neural-style-transfer-app

Install Dependencies

pip install -r requirements.txt

The requirements.txt file should include:

torch
torchvision
streamlit
pillow
requests

Usage

Running the Streamlit App

streamlit run app.py

After running the command, your web browser should automatically open with the app running at http://localhost:8501.

Using the App

1. Upload a content image (the structure you want to preserve)
2. Upload a style image (the artistic style you want to apply)
3. Choose between low or high resolution output
4. Click the "Stylize!" button to begin the process
5. Wait for the algorithm to complete (this may take several minutes)
6. View and download the stylized result

Implementation Details

The core style transfer algorithm is implemented in style_transfer.py, which uses:

• VGG19 pre-trained network for feature extraction
• L-BFGS optimizer for image transformation
• Gram matrix computation for style representation
• Multi-resolution approach for higher quality results

Acknowledgements

This implementation is based on the Neural Style Transfer technique introduced by:

• Leon A. Gatys, Alexander S. Ecker, and Matthias Bethge in their paper "A Neural Algorithm of Artistic Style" (2015)

License

MIT License

Contributing

Contributions, issues, and feature requests are welcome! Feel free to check the issues page.