@citnaj any thoughts on using unets for generative art?
-
-
Replying to @ajankelo
Can you expand on that question? “Generative” as far as I understand it is meant to imply from scratch (starting with random vector as a seed) so if that’s the intent then Unets don’t make sense. But I’m not sure if that’s what you’re asking.
1 reply 0 retweets 0 likes -
Replying to @citnaj
I am thinking along the lines of creating, let’s say a portrait of a fake person based on real portraits, or works of art using existing images as a seed.
1 reply 0 retweets 0 likes -
Replying to @ajankelo
Really that's an image to image transformation you're describing there then.... Which of course Unets are great at. But the real question here is how do you define the transformation that you're training on. The Unet part is the no-brainer relatively speaking.
1 reply 0 retweets 0 likes -
Replying to @citnaj
Interesting. Do you think the superres/gan nb is a good place to start from for this? I also to read/watch more on image to image transformation
1 reply 0 retweets 0 likes
You mean for the fastai course? Yeah I think that's a good start.
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.