News for hackers. Stuff that matters.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by Mr_P on Monday April 17, 2017 @12:44AM (#14129259)
That animated gif is ridiculously misleading. To someone who doesn't know what's going on, it would look like this can hallucinate Emma Watson's face out of a 5x5 px image.

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).

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by dmos62 on Monday April 17, 2017 @06:04AM (#14130067)
It is impressive given current state-of-art. Your summary doesn't seem fair. Take a look at the first set (Set5) [1], and set upsampling scale to 4x and flip between Bicubic and LapSRN. What seems most exciting to me, is that LapSRN is more resource efficient than bicubic.

[1] http://vllab1.ucmerced.edu/~wlai24/LapSRN/results/Set5/

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by x1798DE on Monday April 17, 2017 @08:43AM (#14130520)
Whether or not their sampling is impressive given the current state of the art (though in that same set the other, non-bicubic methods seem to do pretty well compared to LapSRN), that gif definitely seems misleading. It's not obvious at all what is happening here, and it doesn't really even show off their method very well (as, say, separate gifs for each of the downsampling rates and/or clearly labeled frames). The way it is now, it really seems like each progressive enhancement is another stage in the image reconstruction.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by dmos62 on Monday April 17, 2017 @05:45PM (#14134691)
The banner gif is confusing, reminds me of the Super Troopers Enhance scene. Not fitting for an academic publication.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by thanatropism on Monday April 17, 2017 @11:15AM (#14131481)
I remember a proprietary image format (complete with a proprietary converter and viewer for Windows) in the 1990s that dubbed itself a "fractal compression algorithm" -- and did an impressive job with things like water ripples in rivers or grassy terrain.

You could "zoom in" 8-10X in an original nature photograph and be semi-convinced.

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by vernie on Monday April 17, 2017 @12:11PM (#14131991)
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by petters on Monday April 17, 2017 @01:57AM (#14129438)
When I was still in computer vision, "super-resolution" meant taking multiple images and combining them to really be able to see what was in the scene.

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.

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by BugsJustFindMe on Monday April 17, 2017 @03:32AM (#14129687)
I don't know when you were in computer vision, but when I was in computer vision a bit more than a decade ago it meant both.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by yousry on Monday April 17, 2017 @03:40AM (#14129703)
I would also call it a perverted use of the term super-resolution. NNs perform image manipulations beyond resolution enhancement. They add features based on trained sources.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by BugsJustFindMe on Monday April 17, 2017 @03:49AM (#14129728)
And? Do a search on scholar for "Example-based Super Resolution" or "Markov Networks for Superresolution". The term has been used this way in literature for maybe 20 years now.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by petters on Monday April 17, 2017 @04:20AM (#14129813)
OK. In my (limited) view of the field these kind of things started with "Super-Resolution From a Single Image" in 2009 but apparently it is older than that.

To me personally, it feels weird to call this super-resolution, but I realize I may be in the minority.

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by posterboy on Monday April 17, 2017 @04:33AM (#14129857)
Isn't either essentially the same. To say guessing neural networks combine the input of multiple images would be an understatement.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by ISL on Monday April 17, 2017 @10:06AM (#14130941)
One is attempting to ascertain an underlying commonality given multiple measurements, the other is attempting to infer more information than is present in any measurement.

They're quite different.

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by petters on Monday April 17, 2017 @02:36PM (#14133288)
Have you seen the "super-resolution" results with faces and NN?

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.

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by amelius on Monday April 17, 2017 @08:38AM (#14130503)
The results look more blurry than necessary, and in some instances they look too sharp.

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.

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by TuringNYC on Saturday April 15, 2017 @06:27AM (#14120168)
Reminds me of Bourne-style "enhance" feature. Too bad the GIF on the main page appears to be marketing (seems impossible from a single frame like that) and takes away from the achievements of the paper.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by th0ma5 on Monday April 17, 2017 @12:10AM (#14129172)
Yes, that image is deliberately misleading to the lay person... it is showing how it up-samples from various inputs, each having greater resolution. It is impressive, but it seems to imply that it is doing something magical to produce a full resolution picture from a few mere pixels. What is actually doing is impressive enough, but it seems like an odd choice of promoting the work given the obvious visual implication which can't be an accident.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by rini17 on Monday April 17, 2017 @02:26AM (#14129531)
Something like every odd frame is input, every even frame is upsampled output?
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by breeze_em_out on Monday April 17, 2017 @04:54AM (#14129901)
That gif is not even close to how it actually works, that's such a dirty thing to do, that's no accident...
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by bjornsing on Monday April 17, 2017 @04:32AM (#14129854)
Am I missing something, or where are the originals?

(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.

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by dmos62 on Monday April 17, 2017 @06:08AM (#14130075)
The example sets are clickable. That's where the As are.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by bjornsing on Monday April 17, 2017 @07:18AM (#14130258)
As far as I can see there's only other kinds of Bs there (B = cubic(A), B = some_other_super_resolution_menthod(A), etc, etc). Can't find the real As... :P
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by MagnumOpus on Monday April 17, 2017 @08:09AM (#14130386)
The left-most item in the examples is labeled "ground truth". That is your "A" unless we misunderstand you.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by x1798DE on Monday April 17, 2017 @08:47AM (#14130536)
I think that the parent means that the "Ground Truth" is the high resolution original, and they never show a "low resolution" original.

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).

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by bjornsing on Monday April 17, 2017 @03:12PM (#14133603)
Yes, that's what I meant!

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

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by ClassyJacket on Monday April 17, 2017 @11:42AM (#14131719)
They don't appear to show the input anywhere.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by ghusbands on Monday April 17, 2017 @10:38AM (#14131165)
There are signs in the data that the downsampling technique that is being used is not gamma-correct. That would somewhat undermine the results (and also the NNs, if they were trained on similarly broken inputs). Can one of the authors clarify that gamma-correct downsampling/blurring/convolution was used?
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by vanderZwan on Monday April 17, 2017 @11:45AM (#14131735)
With this idea of NNS to predict how to upscale images, I'm kind of wondering if LenPEG isn't becoming more relevant than ever:

http://www.dangermouse.net/esoteric/lenpeg.html

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by agumonkey on Monday April 17, 2017 @04:10AM (#14129784)
Please be sure to leave time to start loading the various filtered versions, if you hover too fast the alternate image won't load, silently, and you'll be left with the last successfully loaded one.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by ghusbands on Monday April 17, 2017 @10:35AM (#14131142)
Almost every site like this gets this wrong - if the switching only happens once the image has finished loading, you need to otherwise hide the now-wrong image. Having to second-guess whether an image has loaded is asinine.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by agumonkey on Monday April 17, 2017 @12:03PM (#14131914)
some have a slightly different UX making it more obvious, IIRC fabrice bellard site for BPG format uses mouse down events to show the filtered version, it's way random than hovering on a link so you intuitively wait more on the first held down.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by Folkloretiger on Monday April 17, 2017 @05:51AM (#14130038)
Could this be applied to cosmological images from large telescopes.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by semi-extrinsic on Monday April 17, 2017 @06:02AM (#14130063)
In the trivial sense, yes of course it could be applied. The real question is "would super-resolution be useful for images from large telescopes?", and I believe the answer is "mostly no".

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.

Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by deepnotderp on Monday April 17, 2017 @12:09AM (#14129169)
Before the inevitable "enhance" comments come in, please note that these nets are making up the information to insert based on information from the training set.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by goombastic on Monday April 17, 2017 @12:47AM (#14129265)
I wonder what this can do for astronomy.
Re:Deep Laplacian Pyramid Networks for Fast and Accurate Sup... by hobofan on Monday April 17, 2017 @03:45AM (#14129717)
How do you mean? AFAIK since those networks "halucinate" the additional data based on similar but actually unrelated data, its not of much use in science where accurate data is important.