[HN Gopher] High Fidelity Image Generation Using Diffusion Models
___________________________________________________________________
High Fidelity Image Generation Using Diffusion Models
Author : theafh
Score : 32 points
Date : 2021-07-16 17:40 UTC (5 hours ago)
(HTM) web link (ai.googleblog.com)
(TXT) w3m dump (ai.googleblog.com)
| antihero wrote:
| I wish they'd show the outputs next to the source, not just the
| input. They've created something convincing looking, but how true
| is it to what was actually there?
| edge17 wrote:
| Sometimes it doesn't matter. Thats why sometimes lossy
| compression is fine.
| antihero wrote:
| It's important to gauge the value of the result - is it
| producing something that looks nice or something useful.
| 317070 wrote:
| It just looks nice. Once you downsample, you cannot
| accurately upsample again. If the upsampled image looks
| real to you, and if you would downsample that again you get
| the same downsampled image, then you did the best you
| could.
| gigatexal wrote:
| isn't that what they do here? https://iterative-
| refinement.github.io/assets/cascade_movie2...
| jaschasd wrote:
| They show the source images next to the super-resolution output
| on the website for the iterative refinement paper:
| https://iterative-refinement.github.io/
| antihero wrote:
| Thanks!
| barryp wrote:
| Yes, I'd like to see 3 images: A) the original image _before_
| it was downscaled to 64x64, B) the 64x64 input to the upscaler,
| and C) the final upscaled image. So we can compare A and C
| fxtentacle wrote:
| This appears to be class-conditioned, meaning it only works for
| images that show objects in one of the 1000 imagenet categories.
___________________________________________________________________
(page generated 2021-07-16 23:01 UTC)