Chapter 4: Enhancing and Styling Images with DeepDream, Neural Style Transfer, and Image Super-Resolution
Although deep neural networks excel in traditional computer vision tasks for purely practical applications, they have a fun side too! As we'll discover in this chapter, we can unlock the artistic side of deep learning with the help of a little bit of cleverness and math, of course!
We'll start this chapter by covering DeepDream, an algorithm used to make neural networks produce dream-like images. Next, we'll seize the power of transfer learning to apply the style of famous paintings to our own images (this is known as Neural Style Transfer). Finally, we'll close with Image Super-Resolution, a deep learning approach that's used to improve ...
Get TensorFlow 2.0 Computer Vision Cookbook now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
