Re:Stable Diffusion 2.0 Release
by xnyan on Wednesday November 23, 2022 @10:33PM
(#33727542)
When our c suite decides on an ad campaign and tells our artists to draw normal humans, those people have 3 legs or upside down teeth exactly 0% of the time. Humans have many many limitations, but with every model I’ve tested there’s a set of errors that would virtually never be made by any human.
[ Parent
| on HN ]
> with every model I’ve tested there’s a set of errors that would virtually never be made by any human
I guess you've never seen my drawings...
Re:Stable Diffusion 2.0 Release
by vintermann on Thursday November 24, 2022 @05:05AM
(#33729565)
I think it's interesting that drawing too many fingers is a mistake kids make, too, although with less photorealism otherwise. I guess there's a reason all thosr famous artists drew hundreds of hands as practice as well.
I agree. Cars break down and crash, they'll never replace horses.
I think this analogy doesn't hold water - horses aren't exactly a beacon of reliability (having owned one).
I've already seen tools that support workflows where you compose art by iteratively generating a piece of it, performing some correction, and repeating. So, I think there's room in the art world for less than perfectly generated art. That said, let's not kid ourselves that the typical failure modality of ML today (99% correct enough, 1% disastrously incorrect) doesn't either cause it to be entirely useless in many applications or end up wreaking havoc on end users in others.
It's only an analogy, but it serves to underscore the last point you make. Initial versions of the technology can make some genuine horrors but you're blinding yourself to progress if you can't see the potential in it.
Re:Stable Diffusion 2.0 Release
by sussmannbaka on Thursday November 24, 2022 @02:39AM
(#33728709)
the cars we're talking about here have a random amount of wheels and sometimes morph into cosmic horrors mid-ride.
Re:Stable Diffusion 2.0 Release
by iceburgcrm on Thursday November 24, 2022 @02:14AM
(#33728567)
In fairness ad campaigns have people who review the creative. Now I could see a band using an image like that or something edgy
Re:Stable Diffusion 2.0 Release
by kennyloginz on Thursday November 24, 2022 @02:28AM
(#33728647)
As an outsider, this rings true to me. I still don’t see any reduction of hours involved in producing professional level works. Generating YouTube thumbnails, sure.
Re:Stable Diffusion 2.0 Release
by boardwaalk on Thursday November 24, 2022 @04:13AM
(#33729227)
No one is suggesting anyone take the first image out of a model without any human-based filtering.