Live data from Hacker News

Spent $15 in DALL·E 2 credits creating this AI image

pub.towardsai.net

101–110 of 153 posts

Re: Spent $15 in DALL·E 2 credits creating this AI image

#102
If you think it’s hard to get an AI to render what’s in your mind, try another human artist. Specifying something visually complex with an assumption that it’ll be precisely what you’re imagining is shockingly hard. I’m not surprised prompt creation is so complex. At least with the AI bots the turn around time for iteration is tight. That said humans likely iterate fewer times, but each iteration takes a long time.

Re: Spent $15 in DALL·E 2 credits creating this AI image

#103
My two cents: the techniques OP uses are absolutely valid, but I've found much more success "sampling" styles and poses from existing works.

Rather than trying to perfectly describe my image, I like to use references where the source material has what you want. With minimal direction these prompts get impressively close:

"larry bird as a llama, dramatic basketball dunk in a bright arena, low angle action shot, from the movie Madagascar (2005)" https://labs.openai.com/s/wxbIbXa0HRwwGUqQaKSLtzmR

"Michael Jordan as a llama dunking a basketball, Space Jam (1996)" https://labs.openai.com/s/mX4T5Iak8CMO1rPAmjRb7oyH

At this point I'd experiment with more stylized/recognizable references or add a couple "effects" to polish up the results.

Re: Spent $15 in DALL·E 2 credits creating this AI image

#104
post #89

Earlier quoted context omitted.

Quoted post unavailable.

I suspect this is a joke, but I did find that it was a little overzealous with the filtering. I was trying to get someone (not a specific person) shouting or with an angry expression, and a few prompts I came up with were blocked. Not banned though.

I kept getting a scene with "two people holding hands" blocked, it allowed "two people kissing" and then when I tried "and wife" instead of "two people" it banned me. (They unbanned me when I emailed them though.)

Oddly, the ones it blocked were more sfw than several others it allowed, but of course I don’t know what the outputs would’ve been…

Re: Spent $15 in DALL·E 2 credits creating this AI image

#106
post #34
post #21

Earlier quoted context omitted.

Wow, I've had the same trouble and these are some great tips! Thanks for sharing

Anytime! I have uploaded the image in question: the initial prompt with first generated images, the extended raw image, and then the one with the added details on the city. https://imgur.com/a/QEU7EJ2

This is a fantastic end result. Thanks for sharing your process to get there.

Re: Spent $15 in DALL·E 2 credits creating this AI image

#107

>the ball is positioned in such a way that the llama has no real hope of making the shot I love that we're at the level where the physical "realism" of correctly representing quadrupedals playing basketball is a thing now. I suppose the next level AI will be expected to model a full 3d environment with physical assumptions based on the prompt and then run the simulation

That's the only way to get reliably usable output. There's a lot of "80% there but not quite" in the current version, which makes it more of a novelty than a useful content generator. The problem with moving to 3D is there are no almost no 3D data sources that combine textures, poses (where relevant), lighting, 3D geometry and (ideally) physics. They can be inferred to some extent from 2D sources. But not reliably. H…

While not fully. There is a lot of freely available 3d models that can used as a starting point. Id love a dalle2 for 3d model generation. Even if no texture lighting physics was there.

Re: Spent $15 in DALL·E 2 credits creating this AI image

#108
post #29

Earlier quoted context omitted.

When I see some of the bad pictures it produces I think we are nowhere near AGI

Most people would draw even worse pictures given the same prompts.

most neural networks would draw even worse pictures given the same prompts

Re: Spent $15 in DALL·E 2 credits creating this AI image

#109

I was curious to compare results with Craiyon.ai Here is "llama in a jersey dunking a basketball like Michael Jordan, shot from below, tilted frame, 35°, Dutch angle, extreme long shot, high detail, dramatic backlighting, epic, digital art": https://imgur.com/a/7LoAtRx Here is "Llama in a jersey dunking a basketball like Michael Jordan, screenshots from the Miyazaki anime movie", much worst: https://imgur.com/a/g99G7…

Fascinating, are there any other similar products in this same category as DALL.E and Craiyon?

Re: Spent $15 in DALL·E 2 credits creating this AI image

#110

I was curious to compare results with Craiyon.ai Here is "llama in a jersey dunking a basketball like Michael Jordan, shot from below, tilted frame, 35°, Dutch angle, extreme long shot, high detail, dramatic backlighting, epic, digital art": https://imgur.com/a/7LoAtRx Here is "Llama in a jersey dunking a basketball like Michael Jordan, screenshots from the Miyazaki anime movie", much worst: https://imgur.com/a/g99G7…

Fascinating, are there any other similar products in this same category as DALL.E and Craiyon?

wombo.ai and midjourney
Post reply on HN