[maxander]

So, basically, this is the thing in a crime detective movie where the forensic analyst is looking at a terrible pixelated surveillance camera still and says "enhance," and the computer magically increases the resolution to reveal the culprit's face.

Just another entry on the "things that are supposed to be impossible that convolutional nets can do now."

[Eliezer]

And that's how that guy whose face appeared a few times in ImageNet became the world's most wanted terrorist, on the run for thousands of crimes.

[nullc]

Ding ding.

But good luck convincing a jury that a maximum likelihood decode from a few grainy pixels wasn't reliable when it gave a crystal clear output.

[nerdponx]

http://www.slate.com/articles/news_and_politics/jurisprudenc...

[taneq]

I bet he's hiding out with that lab tech whose poor technique lead to their DNA being in hundreds of crime scene samples.

[eejr]

yup, to certain point! there are information theoretic limits though. You can fill in information, but there will be biases to a certain point. in this case defined by the dataset. if the "enhance" is too strong, we should be careful with what we do with the results in forensics.

but man, it can make your internet pics look smooth! :) thanks for the comment!

[paulsutter]

If you have multiple images of the same scene (for example, from video frames), you should be able to use information across frames for a true enhancement?

[CamperBob2]

Yes, and it's almost ridiculous how well that can work:

https://www.youtube.com/watch?v=ONZcjs1Pjmk

[eejr]

;) sshh! don't say that too loud yet. but, remember our names...

[microcolonel]

So you're saying this could be useful for stereo imagery and video? ;- )

[nitrogen]

This (overenhancement) was a minor plot point in Crichton's novel Congo, IIRC.

[dankohn1]

Here is the ridiculous Let's Enhance supercut that all just became realistic: https://www.youtube.com/watch?v=LhF_56SxrGk

(Created by the super talented duncanrobson)

[duaneb]

I imagine you'd want PII erased from the training set, but the danger stands.

[joelthelion]

Except it might unblur to the face of someone else.