Part 2: Getting Hands-On with the Deepfake Process

This part of the book is all about getting hands-on with the code. We will look deep into exactly what it takes to make a deepfake from beginning to end, leaving no stone unturned or line of code unexplained. If you’re here for the code, this is the section for you.

In the first chapter of this section, we’ll examine extraction. This is the process of getting all the faces out of a video so that we can use them in other stages of the process. We’ll look at the process of turning a video into frame images, then we’ll go through all the code necessary to turn the frames into clean, aligned faces with matching mask images ready for training. After that, we’ll look into training, examine the neural network from the bottom up, and then show the entire learning process of the model. Finally, we’ll get into conversion, where we’ll examine the process of going through every image to swap...