News for hackers. Stuff that matters.

A comment on Generative Adversarial Networks for Extreme Learned Image Compression. Parent comment

Re:Generative Adversarial Networks for Extreme Learned Image... by make3 on Wednesday April 11, 2018 @10:07PM (#16817447)
It's not as bad as you say. Sure, the positions of the individual leaves and of the grain of the concrete and of little puffs of the clouds are hallucinated, but most salient semantic features like the presence of a car or of a person are left untouched.

In other words, mostly only unimportant details are hallucinated, which is what we want.

[ Parent | on HN ]
Re:Generative Adversarial Networks for Extreme Learned Image... by rasz on Friday April 13, 2018 @03:11PM (#16833052)
Car behind the bus is gone, bus license plate got some fake text. Its pretty bad as far as reproducing original.
[ Parent | on HN ]