3 ms·
Super resolution can and does work in some circumstances. By introducing a new source of information (the memory of the NN) it can reconstruct things it has se
by nl 6y ago
Super resolution can and does work in some circumstances.
By introducing a new source of information (the memory of the NN) it can reconstruct things it has seen before, and generalise this to new data.
In some cases this means hallucinations, true. But in other times (eg text where the NN has seen the font) it is reconstructing what that font is from memory.
- panta 6y agoRecalling memory of something different and using it to interpolate is a form of hallucination.
- nl 6y agoNot if it’s written text and you are selecting between 26 different letters. It’s a probabilistic reconstruction, but that’s very different to a hallucination.
- erichocean 6y agoWait until panta discovers image compression…
- panta 6y agoLossy image compression creates artifacts, which are in a way of form of falsely reconstructed information - information which wasn't there in the original image. Lossless compression algorithms work by reducing redundancy, but don't create information where it wasn't there (thus being very different from super-resolution algorithms).
- cycomanic 6y agoBut the thing is, in that case the information contained in the images was actually much less than what we are meant to make believe. So if we are reconstructing letters from a known font we essentially are extracting 8 bits of information from the image. I'm pretty certain that if you distort the image to an SNR equivalent of below 8 bits you will not be able to extract the information.