Saw this in Tumblr
QuoteA research team trained a machine learning algorithm (a GAN they called AttnGAN) to generate pictures based on written descriptions. It?s like Visual Chatbot in reverse. When it was just trained to generate pictures of birds, it did pretty well, actually.
(https://78.media.tumblr.com/d4605122d4139397ebb47d69aedb48a7/tumblr_inline_pdk61rZjAc1rl9zu7_540.png)
It does a horrifying job with humans because it can never quite seem to get the number of orifices correct.
(https://78.media.tumblr.com/0b39505309e5b6d7589d1d4a98c5b7e0/tumblr_inline_pdjdg1jIPw1rl9zu7_540.png)
From asking it to draw *a* sheep. In that case, it knows exactly what to do. It draws the sheep, and then just to be safe it fills the entire planet with wool too.
(https://78.media.tumblr.com/143c4aad68412e23c802462f3b2766ec/tumblr_inline_pdjdroce7U1rl9zu7_540.png)
Try it here:
http://t2i.cvalenzuelab.com/
Everything I type in ends up looking creepy :mmyulost:
(http://i66.tinypic.com/6nvwh4.jpg)
Cxxxxxxcccxxxxxxxxzzx CXXXX
This is demonic
Quote from: squid on August 25, 2018, 03:52:36 PM
(http://i66.tinypic.com/6nvwh4.jpg)
cnncncmcnc
C
C
Very ... abstract but real. loving this