Fascinating (and very extensive) writeup on using deep learning to turn website design mockups into actual code:https://blog.floydhub.com/turning-design-mockups-into-code-with-deep-learning/ …
Show this thread
I think in the near future, deep learning will be used to dynamically generate ad creatives adapted to different contexts (tablet, phone, vertical banner, horizontal banner...) from a single generic mockup plus raw assets
-
-
Also generating tons of candidate ideas for designs that designers can select from and iterate on feels plausible in the near future.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Extending the idea, utility of DSLs can be vastly improved with this technique (CLG - Computer Language Generation?)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Alibaba has already developed this technology called Ruban since 2016. The system was designed to accommodate the need of its 11.11 shopping festival. 170 million banners are generated within one day.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Not necessarily generated the banner as code though.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I had the same idea. Banner `targeting` with GANs is very useful. All you need is just one banner with typical man who drinks coffee and a bunch of GANs that can generate same banner but with different persons (woman | older man | people of different ethnic origin)
-
Or different drinks: ( juice | coffee | cocktail ...). Most of banners has a low resolution and this is good for current generation of GANs
End of conversation
New conversation -
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.