News for hackers. Stuff that matters.
Re:Subpixel: A subpixel convolutional neural network impleme... by eejr on Saturday October 01, 2016 @04:05PM (#12619656)
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!

Re:Subpixel: A subpixel convolutional neural network impleme... by paulsutter on Saturday October 01, 2016 @04:26PM (#12619729)
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?
Re:Subpixel: A subpixel convolutional neural network impleme... by CamperBob2 on Saturday October 01, 2016 @10:14PM (#12620785)
Yes, and it's almost ridiculous how well that can work:

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

Re:Subpixel: A subpixel convolutional neural network impleme... by eejr on Saturday October 01, 2016 @04:45PM (#12619808)
;) sshh! don't say that too loud yet. but, remember our names...
Re:Subpixel: A subpixel convolutional neural network impleme... by microcolonel on Saturday October 01, 2016 @05:22PM (#12619945)
So you're saying this could be useful for stereo imagery and video? ;- )
Re:Subpixel: A subpixel convolutional neural network impleme... by nitrogen on Saturday October 01, 2016 @08:57PM (#12620582)
This (overenhancement) was a minor plot point in Crichton's novel Congo, IIRC.