take a look at "Figure 8" from the original paper:
we can smoothly interpolate between faces, so it seems impossible to me that these are just memorised from the training set
we can smoothly interpolate between faces, so it seems impossible to me that these are just memorised from the training set
However, aligned faces are definitely not required - I didn't do any kind of alignment for my anime faces and you can see the eyes/nose/mouth in all sorts of positions in the samples & videos.