Key feature set
The human face and body are key entities in the task of fake content generation. While deep learning architectures usually do not require hand-crafted features, a little nudge goes a long way when complex entities are involved. Particularly when dealing with the human face, apart from detecting the overall face in a given image/video, a deepfake solution also needs to focus on the eyes, mouth, and so on. We discussed different modes of operation in the previous section, where we highlighted the importance of different sections of a face and their impact on improving the believability of generated fake content.
Facial landmarks
Facial landmarks are a list of important facial features such as the nose, eyebrows, mouth, the contours of the eyes, and so on. The goal is the detection of these key features using some form of a regression model. The most common method is to leverage a predefined set of positions on the face or body that can be efficiently tracked...