Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution
vllab1.ucmerced.edu
vllab1.ucmerced.edu
Instead, this is essentially a fancy 2x upsampling, and the gif shows every other frame as a super-resolution result. In fact, if you look at details like her eye, it's not even doing that great of a result (unsure if this is supposed to be impressive given the current state-of-the-art).
You could "zoom in" 8-10X in an original nature photograph and be semi-convinced.
Now it often means: take a single image and have the computer guess/make up image content. Not the same at all, IMO. There are worse examples than this, though.
To me personally, it feels weird to call this super-resolution, but I realize I may be in the minority.
They're quite different.
Imagine grainy CCTV footage as evidence in court. A serious image analyst may attempt using super-resolution to obtain the face of a suspect.
But a NN could happily insert a very plausible face that looks great. It could have little to do with reality, though.
I think deep learning is a more promising approach, as for superresolution you really need to "invent" missing pieces of the image, and this can be highly context dependent.
(I'm assuming "super-resolution" in this context is like a function B = super_resolution(A). I can see lots of what I think are Bs, but where are the As? Aren't they super-relevant?)
() No pun intended.
Presumably the process goes:
High res. (A') --(Downsample)--> Low res. (A) -- (Upscale) --> Reconstructed (B)
The sets show A' and B, but not A (unless I'm way off on what's happening in this process).
If the Ground Truth where the A then I'd implement my very own "optimal super-resolution" like this:
def optimal_super_resolution(A): return A
It would be a huge breakthrough. :P
I guess (one of) the exception(s) would be if you have many smaller telescopes that could scan the sky much faster than the few big ones, the small ones could use super-resolution techniques to look for objects that the big ones might find interesting. But I think "astronomer time" rather than "telescope time" is usually the limiting factor.
Mind you, telescopes already do a host of physical tricks to improve resolution, like sensor cooling, stacking images, adaptive optics with laser guide stars, advanced noise filtering etc. Whether ML super-resolution stuff would actually be useful on top of all those tricks is a question for the astronomers in the crowd.