Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I recreated famous album covers with DALL-E

lucytalksdata.com

Post reply on HN
Print view View original

101–110 of 137 posts

Re: I recreated famous album covers with DALL-E

#101

Post by DonaldFisk »

fifilura wrote: post #45
DubiousPusher wrote: post #20

Earlier quoted context omitted.

I actually thought there were some interesting things revealed here. Particularly interesting to me were the Nevermind and Abbey Road results. Nevermind because it showed this weakness in the model in understanding what a dollar bill is. The most novel result being the image where the baby's visage appears inside the dollar. Regarding Abbey Road, I found it interesting that the model's concept of a public person span…

For me, what was interesting to see was that it didn't manage very well with the Abbey Road question. I had imagined that some parts of the training data would consist of the actual image, and it would find a good match for it somewhere deep into it's artificial consciousness.

Maybe it would have done better if the prompt had been "Four men - George Harrison, Paul McCartney, Ringo Starr and John Lennon - striding across a zebra crossing situated on Abbey Road, outside EMI studios in London."

Re: I recreated famous album covers with DALL-E

#102

Post by joshschreuder »

whirlwin wrote: post #62
powersnail wrote: post #35

DALL-E is still highly probabilistic in its judgement. For instance, in this article, it keeps putting "fire" in the the background on something that is likely to be on fire, rather than lighting up the person. I have a similar experience. In my own experiment, I can't get DALL-E to turn off the street lamp at a bus stop in the darkness. I've tried "no light", "broken street lamp", etc.; no use. Any mention of "stree…

I have experienced violent or harmful settings to be avoided by DALL-E. E.g. setting a person on fire. Same with drowning - seems to be impossible/hard to generate

I asked for "The Walking Dead directed by x" and got a content violation, I guess because my prompt included "dead".

Re: I recreated famous album covers with DALL-E

#103

Post by svnpenn »

twostorytower wrote: post #90
Michelangelo11 wrote: post #6

Man, after seeing Stable Diffusion's output, DALL-E's looks just janky. Like watching a propeller plane after seeing a jet. Crazy how fast the tech is moving.

DALL-E is capable of very high quality photorealistic images with the right prompts. Here's one I made: https://imgur.com/yAzKkHb “High detail, macro portrait photo, a handsome Australian man with a strong jaw line, blue eyes and brown hair, smiles at the camera, set in an outdoor pub at golden hour, shot using a ZEISS Supreme Prime lens”

> shot using a ZEISS Supreme Prime lens

that seems too specific. You wouldn't even request that in real life, unless you were trying to be ironically pretentious.

Re: I recreated famous album covers with DALL-E

#105

Post by sdflhasjd »

svnpenn wrote: post #103
twostorytower wrote: post #90

Earlier quoted context omitted.

DALL-E is capable of very high quality photorealistic images with the right prompts. Here's one I made: https://imgur.com/yAzKkHb “High detail, macro portrait photo, a handsome Australian man with a strong jaw line, blue eyes and brown hair, smiles at the camera, set in an outdoor pub at golden hour, shot using a ZEISS Supreme Prime lens”

> shot using a ZEISS Supreme Prime lens that seems too specific. You wouldn't even request that in real life, unless you were trying to be ironically pretentious.

DALLE is trained on labelled data, if a photographer uses these terms to describe their work, using these in the prompts gently herds the image generation towards that kind of media.

Re: I recreated famous album covers with DALL-E

#106

Post by chihuahua »

svnpenn wrote: post #103
twostorytower wrote: post #90

Earlier quoted context omitted.

DALL-E is capable of very high quality photorealistic images with the right prompts. Here's one I made: https://imgur.com/yAzKkHb “High detail, macro portrait photo, a handsome Australian man with a strong jaw line, blue eyes and brown hair, smiles at the camera, set in an outdoor pub at golden hour, shot using a ZEISS Supreme Prime lens”

> shot using a ZEISS Supreme Prime lens that seems too specific. You wouldn't even request that in real life, unless you were trying to be ironically pretentious.

"shot using a ZEISS lens, by a hipster with an ironic mustache"

Re: I recreated famous album covers with DALL-E

#107

Post by j79 »

CamperBob2 wrote: post #92
randymy wrote: post #66

Worth noting that DALL-E automatically “rejects attempts to create the likeness of any public figures, including celebrities". So, you wouldn't be able to get an image that included the 4 Liverpudlians. It does allow you to create fake faces. Might be fun to try and recreate Miles Davis Tutu, Aladdin Sane, Piano Man.

It will yell at you and threaten your account with termination if you try to create anything based on a living person's face, from what I can tell.

Yep. I once tried to create a cartoon dinosaur with hair in the “style” of an ex President (yellow and combed forward), and was warned with a potential ban.

Re: I recreated famous album covers with DALL-E

#109

Post by jasomill »

chihuahua wrote: post #106
svnpenn wrote: post #103

Earlier quoted context omitted.

> shot using a ZEISS Supreme Prime lens that seems too specific. You wouldn't even request that in real life, unless you were trying to be ironically pretentious.

"shot using a ZEISS lens, by a hipster with an ironic mustache"

Supreme Primes are modern, $20,000+ PL mount cine lenses, so probably not popular even among trust fund hipsters for shooting stills.

To be fair, if I had the funds, I'd certainly consider it, though I consider myself slightly mental in my choice of photography equipment.

Re: I recreated famous album covers with DALL-E

#110

Post by chihuahua »

avian wrote: post #59
xwdv wrote: post #7

Although AI artists will destroy a lot of jobs, it will also create demand for new jobs for people who specialize in “paint overs” – taking a high concept output created by AI artists and touching it up to perfection. Or perhaps even beyond just a paint over, and into the realm of recreating an entire AI artwork but with a human touch to get details just right. Looking forward to it.

Here's another spin on this: Imagine you're a software developer. In the near future, your manager wants a feature implemented in your company's app. He throws together a short mail with requirements and sends it to the prompt engineering department. The prompt engineers fix a few typos, clean up the grammar and pepper a few secret sauce keywords around like "in the style of firefox", "in the style of kde". This get…

That sounds better than my current job where I try to clean up other people's shitty code with atrocious dynamic typing, until the type checker is happy with it.
Post reply on HN