Image Style Transfer

Neural image style transfer

Overview

‚ÄčNeural Image style transfer using a pre-trained model given a style and content image to generate a new-style transferred image

Architecture

The application built based on the following nodes:

Image Input node

Two image input to upload your content and style images.

Image to Array node

Convert image to array

Image style transfer node

Apply for image style transfer on a content image using Very Deep Convolutional Networks for Large-Scale Visual Recognition (VGG-19) weights.

Image preview node

Show output image on the output tab

Sample Run

  • Drag Content image into the first image input and the style on the second input image nodes.

  • Set the maxIterations input field, to improve the quality of style transfer you have to increase the maximum iterations field.

Execution time increases based on the max iterations

  • Click on execute app

  • Click on the image preview node and see the result on the output tab

Sample Run - Form

If you want to run any application without going into the app details, Baseet.ai offers this option via Run an Example form in the overview section in the app details page.

For this app, the required inputs are:

  • First input is the Content image that you would like to apply style transfer on it.

  • Second input is the style image that you would like to use in transferring the content image.

Click on Run Example. After the app runs successfully, the output section will be shown as follows:

To enlarge the image click on the image.