Re-enactment using Pix2Pix
Re-enactment is another mode of operation for the deepfakes setup. It is supposedly better at generating believable fake content compared to the replacement mode. In earlier sections, we discussed different techniques used to perform re-enactment, i.e., by focusing on gaze, expressions, the mouth, and so on.
We also discussed image-to-image translation architectures in Chapter 5. Particularly, we discussed in detail how a pix2pix GAN is a powerful architecture that enables paired translation tasks. In this section, we will leverage a pix2pix GAN to develop a face re-enactment setup from scratch. We will work towards building a network where we can use our own face, mouth, and expressions to control Barack Obama’s (the former US president) face. We will go through each and every step, starting right from preparing the dataset, to defining the Pix2Pix architecture, to finally generating the output re-enactment. Let’s get started.
