This an implementation of Neural Style Transfer using Tensorflow. This algorithm was created by Gatys et al. (2015). The paper is here.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.
You will need the following packages:
You can install Python dependencies using pip install -r requirements.txt
,
and it should just work. If you want to install the packages manually, follow the above list.
For Jupyter Notebook, install Anaconda.
Other dependancies are there in the nst_utils.py file.
Download the Pre-trained VGG network and place it in the pretrained-model folder.I have not uploaded the pre-trained VGG model here.
Replace the images/image.jpg and images/style.jpg by your desired content image and styling image respectively.
Run the notebook and get started!:)
Note: I have resized both the original image and the styled image to 600 X 800. Please choose both images greater than or equal to the dimensions specified to get optimum results. Otherwise, change the basewidth, hsize parameters in the notebook and IMAGE_WIDTH, IMAGE_HEIGHT parameters under class CONFIG
of nst_utils.py
- Coursera- Deep Learning specilization courses, Deeplearning.ai by Andrew Ng.
- Tensorflow official website
- Stack-overflow