orchid.gistnoesis.net
AI
generative image models
SHOW HN2022
40points
14comments
#1,089of 2,570 in 2022
Top 43%of 2022 by points
How HN took it
RoastedHalf or more of the comments are critical
0 positive, 3 neutral or asking, 0 mixed, 5 critical, out of 8 comments about the project. Thread score -0.48.
Execution: 0 for, 5 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Interest in seeing how higher resolution changes results
From 1 comment
I for one would like to see what bizarre and/or impossible things this algorithm gets up to once the resolution improves enough to discern body parts.
excalibur on HN
Objections
Images are blurry and not realistic
From 3 comments
Pictures are blurry, NSFW tag is questionable due to poor image quality. For anyone interested they were inspired by DDM but trained on 2 1080Tis over the weekend. So if you're…
godelski on HNThese are obviously just blurry pictures of vegetables.
morninglight on HNI have browser bookmarks that are more NSFW than that blurry mess, arguably clickbait title.
pengaru on HN
NSFW tag is questionable; images are safe for work
From 2 comments
Pictures are blurry, NSFW tag is questionable due to poor image quality. For anyone interested they were inspired by DDM but trained on 2 1080Tis over the weekend. So if you're…
godelski on HNI’d say this (unfortunately) was safe for work.
bjornsing on HN
Title is clickbait
From 1 comment
I have browser bookmarks that are more NSFW than that blurry mess, arguably clickbait title.
pengaru on HN
Requests
Use lower resolution or latent diffusion models given limited compute
From 2 comments
Pictures are blurry, NSFW tag is questionable due to poor image quality. For anyone interested they were inspired by DDM but trained on 2 1080Tis over the weekend. So if you're…
godelski on HNFYI if you only have a limited amount of compute, latent diffusion models will converge much faster. https://github.com/CompVis/latent-diffusion
fxtentacle on HN
Train on photos rather than videos
From 1 comment
I suspect you'll have much better results training on photos rather than videos. The individual frames of a video are often blurry, poorly composed, and uninteresting, but photos…
Scaevolus on HN
Alternatives named
Smutfinder / driftwheeler neural porn browsing
From 1 comment
Neural nets for porn browsing (not generation): http://driftwheeler.com Article in The Register (more than 5 years ago):…
onetime8192 on HN
Latent diffusion models (CompVis)
From 1 comment
FYI if you only have a limited amount of compute, latent diffusion models will converge much faster. https://github.com/CompVis/latent-diffusion
fxtentacle on HN