Chapter 14. Generating Icons Using Deep Nets
In the previous chapter we looked at generating hand-drawn sketches from the Quick Draw project and digits from the MNIST dataset. In this chapter weâll try three types of networks on a slightly more challenging task: generating icons.
Before we can do any generating we need to get our hands on a set of icons. Searching online for âfree iconsâ results in a lot of hits. None of these are âfree as in speechâ and most of them struggle where it comes to âfree as in beer.â Also, you canât freely reuse the icons, and usually the sites strongly suggest you pay for them after all. So, weâll start with how to download, extract, and process icons into a standard format that we can use in the rest of the chapter.
The first thing weâll try is to train a conditional variational autoencoder on our set of icons. Weâll use the network we ended up with in the previous chapter as a basis, but weâll add some convolutional layers to it to make it perform better since the icon space is so much more complex than that of hand-drawn digits.
The second type of network weâll try is a generative adversarial network. Here weâll train two networks, one to generate icons and another to distinguish between generated icons and real icons. The competition between the two leads to better results.
The third and final type of network weâll try is an RNN. In Chapter 5 we used this to generate texts in a certain style. By reinterpreting icons ...
Get Deep Learning 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.