Which is exactly what your gifs do. Oh, you thought your gifs somehow tracked the facial features like facial recognition does, didn't you? 
These are is images that the Apple face id sees, all these little dots are used to create a depth map of someone's face and since our faces are all unique this form of identification is virtually infallible.

Now to make it easier to comprehend my method of face comparison I've used a face with dots pasted on and these images are an example of the same data the apple software uses to unlock a three dimensional matching image.

In the following gif your fading technique is used and all we get is a jumbled mess.

But by using my methodology we go far beyond your crude fade code and now we are looking at a program which utilizes millions of lines of code to truly unlock the mathematical perfection of a precise rotating match between millions of spatially co-ordinated depth-mapped pixels.

JohnM