I think it might be useful to draw a boundary between dogs, whom humans have selectively bred for a long time including for their ability to communicate with us and vice versa, and other animals. Also, for dogs I’d mostly omit looking at their mouths (except for open/closed and how so) since the mouth shape is mostly fixed. They express more with their eyes and where they direct their attention. And with the rest of their bodies, of course.
I do agree that in general it’s important to distinguish when someone or something just has a face shaped a certain way, vs. when they’re making a facial expression. IRL that might be a matter of movement? As in, we don’t hold a single expression without moving, which is a significant clue.
I’d also draw a boundary between mammals and non-mammals, I might be able to tell a few things from a bird’s behavior, like if it’s afraid, but not from just a picture of its face.
Good points. Though I claim that I do hold the same facial expression for long periods sometimes, if that’s what you mean by ‘not moving’. In particular, sometimes it is very hard for me not to screw up my face in a kind of disgusted frown, especially if it is morning. And sometimes I grin for so long that my face hurts, and I still can’t stop.