Re:Image-to-Image Translation with Conditional Adversarial Nets
by zelpa on Tuesday November 22, 2016 @11:10AM
(#13015022)
I wonder if you were to average the design of the bicycles whether it would actually produce something that works?
[ Parent
| on HN ]
Re:Image-to-Image Translation with Conditional Adversarial Nets
by dTal on Wednesday November 23, 2016 @10:23AM
(#13023208)
I would have thought that, if you are smart enough to find a "bicycle vector space" in which averaging sketches of a bicycle produces another valid sketch of a bicycle, then you probably already know enough about bicycles to design one without the input of imperfect sketches.