pcsingh / Neural_Style_Transfer

Deep learning model for style transfer using VGG19 pre-trained model.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Neural_Style_Transfer

GitHub repo size

This is a Deep Learning model which will help you to transfer an style onto your favourite image. I have used VGG19 pre-trained model which consists of 16 convolution layers, 3 Fully connected layer, 5 MaxPool layers and 1 SoftMax layer. Below you can see the demo of the model.

style image 1     

Output after 1000 iterations

How can you use it?

You can use this model to transfer your favourite style on your content image. I am supposing that you know about the Google Colab. First fork the repo from top right corner. Clone it using your cli tool. Go to the Google Colab or your local Jupyter Notebook and open NST.ipynb file.

Make small changes assign position of your style and content file to variables style and content in the second cell of notebook.

style = 'your_style_image'
content = 'your_content_image'

Hurray! You are ready to see the output of the model. Run all the cells. You can also experiment with the style and content images given in the images folder.

Need help? feel free to contact @ premchandra.singh.5268@gmail.com

About

Deep learning model for style transfer using VGG19 pre-trained model.

License:MIT License


Languages

Language:Jupyter Notebook 100.0%