This is project showcases how we can use a CNN to gather style and content information form an image and then use that to generate another image with same content as one image and same style as the other essentially transfering the style of one image into another.
Table of Contents
This is project showcases how we can use a CNN to gather style and content information form an image and then use that to generate another image with same content as one image and same style as the other essentially transfering the style of one image into another.
To set up a local instance of the application, follow the steps below.
The following dependencies are required to be installed for the project to function properly:
- Python 3.10.16
- Pip 25.1
Now that the environment has been set up and configured to properly compile and run the project, the next step is to install and configure the project locally on your system.
-
Clone the repository
git clone https://github.com/cgs-iitkgp/Neural_Style_Transfer.git
-
Install Python and Pip
- Download installer from: https://www.python.org/downloads/
- Run it and make sure to tick "Add Python to PATH".
- Check in terminal:
python --version pip --version
-
Install Dependencies
pip install -r requirements.txt
-
Running the Program
- Use main_unweighted.py or main_weighted_CNN_layers.py to perform the image style transfer and edit the path of the content as well as style image in the python files to your respective file paths.
python main_unweighted.py python main_weighted_CNN_layers.py
This is an example of the style transfer with the content, style and target image.
The currently active maintainer(s) of this project.
Honoring the original creator(s) and ideator(s) of this project.