Hacker News new | past | comments | ask | show | jobs | submit login

I don't know about that. A lot of the images I've seen SD generate have been interesting but "incorrect" in really important ways that indicate it doesn't _understand_. Things like people's limbs not being connected correctly, or three legs, or a wheel on a car not being in the right place, or elements of perspective that are just off. All things a real artist would never do unless it was intentional.

It's like some of the stuff that GPT-3 generates, it sounds extrememly plausible and realistic but then you read it properly and realise that parts just don't make any sense.

However I don't think these things are show-stoppers, I think we'll see a lot of artists getting good at providing the "feedstock" and guidance to these systems to deal with those quirks and generating some really interesing work.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: