Applying style transfer with TFHub

Implementing Neural Style Transfer from scratch is a demanding task. Fortunately, we can use out-of-the-box solutions that live in TensorFlow Hub (TFHub).

In this recipe, we'll style our own images in just a few lines of code by harnessing the utility and convenience that TFHub provides.

Getting ready

We must install tensorflow-hub. We can do this with just a simple pip command:

$> pip install tensorflow-hub

If you want to access different sample content and style images, please visit this link: https://github.com/PacktPublishing/Tensorflow-2.0-Computer-Vision-Cookbook/tree/master/ch4/recipe5.

Let's take a look at the sample image:

Figure 4.11 – Content image

Let's get started!

How to do it…

Neural Style Transfer with TFHub is a breeze! Follow these steps to complete this recipe:

1. Import the necessary dependencies:

   import matplotlib.pyplot as plt
   import numpy as...