We're going to learn all about facial landmarks in dlib. Before we can run any code, we need to grab some data that's used for facial features themselves. We'll see what these facial features are and exactly what details we're looking for. This is not included with Python dlib distributions, so you will have to download this. We'll go to the dlib.net/files/ site, where you can see all the source code files; scroll to the bottom and you can see the shape_predictor_68_face_landmarks.dat.bz2 file. Click on it and then save it wherever you keep your Jupyter Notebooks for this book.
Okay, so, what exactly is that? What are these 68 landmarks? Well, these landmarks are a common feature set that was generated by training alpha datasets from something called iBUG (https://ibug.doc.ic.ac.uk/resources/facial-point-annotations/), the intelligent...