NVIDIA just released the code: <a href="https://github.com/nvlabs/spade/" rel="nofollow">https://github.com/nvlabs/spade/</a>
I find this ... disquieting. I think its fantastic but I also find something about the <i>lack</i> of uncanny valley troubling.<p>I should feel happier about it, but I can't stop feeling a bit odd that a sketch can go to photorealistic north of the bad so well now: I expected 5-10 more years for this.
Imagine the impact on moviemaking that this will have within just a few iterations of processing power. Thousands of hours of visual effect artist work in film and TV will soon be abstracted into some high-level commands, transformed by software into moving film. Very exciting.
Wow... the effort that went into this article is impressive! Great animations and explanations to decompose what is a pretty complex subject.<p>Nicely done.
That is one of the best articles I have ever seen on a complex technical subject. A reasonable amount of math, great examples, and animations of the process.<p>If the massively online education people had that kind of quality, maybe people would actually finish the courses.
If you want to see the same thing done 18 years ago without new-age machine learning, read <a href="https://www.mrl.nyu.edu/projects/image-analogies/index.html" rel="nofollow">https://www.mrl.nyu.edu/projects/image-analogies/index.html</a> IMO the most elegant vision/graphics algorithm ever written.<p>Specifically this is the "texture-by-numbers" application. Ex: <a href="https://www.mrl.nyu.edu/projects/image-analogies/potomac.html" rel="nofollow">https://www.mrl.nyu.edu/projects/image-analogies/potomac.htm...</a><p>Every single fancypants application of neural nets in graphics today is a retread of one of the applications of the Image Analogies algorithm.