Well… yes… but sometimes you just want the funny misinterpretations.
I don’t think anyone opening this post is actually interested in seeing a bunch of accurate interpretations :)

Another one I completely forgot about, I also have access to the NovelAI image generator, which is completely anime trained. It has a ton of different features that I never use. I just use it as a writing aid for myself, the image generation was kind of tacked on later but grew into it’s own thing.

There’s a mood changer:

Line art (which removes the dollar bill, weirdly enough ) :

And of course inpainting:

There’s also some kind of outpainting, but like I said, I hardly ever use this tool. Here’s a tutorial if you want to read more about it:
https://blog.novelai.net/tutorial-intro-to-image-generation-ef644a1a4885

Oh and it’s completely privacy focused, so images are only saved to your computer.

cc: @M0oP0o@mander.xyz

Midjourney:

Nothing as funny I’m afraid.

✧✨🌿Allo🌿✨✧

@allo@sh.itjust.works

prompt: A cafe at the North Pole named “Opposites”

Vespair

@Vespair@lemm.ee

I’m sure my whimsical entry won’t be divisive.

imgur version in case the lemmy image upload breaks

SomeAmateur

@SomeAmateur@sh.itjust.works

"Yo, I’m the king of the Arctic, top of the frost,

While you’re waddling lost on some South Pole rocks.

I’m the beast of the north, with teeth like knives,

While you slip ‘n’ slide, barely keepin’ alive.

I’m six feet tall, got the muscle and the brawn,

You’re a pint-sized bird, can’t even fly, come on.

They call me the apex, the top of the line,

You’re snack-sized, flappin’—couldn’t handle my grind.

I’m the white ghost, huntin’ seals at night,

While you shiver in groups, tryna huddle tight.

I roam solo, 'cause I run the ice,

You’re just fish food waddlin’—not worth the price."

-ChatGPT

Itrytoblenderrender

@Itrytoblenderrender@lemmy.world

merde alors

@merde@sh.itjust.works

M0oP0o

@M0oP0o@mander.xyz

Well it took over 20 tries in the bingilator. Its not what I wanted but, eh.

“a dollar bill floating in the air with a halo above it and angelic wings. a very small sailor moon is looking at the floating money”

xia

@xia@lemmy.sdf.org

dalle3: ‘A lush overgrowth forest has a vine growing into a skeleton on the ground.’

Thelsim

@thelsim@sh.itjust.works

Prompt: 2000s TV talk show, versus, an orc and an elf sitting across from each other, in a tv talk show studio, the orc is wearing crude metal armor, the elf is wearing flowing white robes, the headline “Opposing views” is writting in bold font at the bottom of the screen, the subtitle “Hobbits, tasty or not?” is written in the bottom left, HD-TV screen shot --ar 91:51 --v 6.1

I actually managed to get the proper subtitle, but I liked this one better. Like Gollum managed to sneak in a quick edit before airing :)

WhatsHerBucket

@WhatsHerBucket@lemmy.world

The OG polar opposites

Starting prompt in dalle: Create an Ornate Yin Yang with dragons on each side and a Chinese culture vibe

Trollivier

@Trollivier@sh.itjust.works

Nightmare VS Day dreaming

Prompt : A portrait split down the middle: the left side of the face is asleep and scared, with nightmarish shadows, terrifying imagery and eerie creatures in the background. The right side of the face is awake and calm, with colorful symbols of peace and creativity like floating flowers, soft clouds, and bright abstract shapes forming a dreamlike, idealistic scene. The two sides blend subtly in the center, showing harmony between fear and peace.

I don’t know, but I felt like sharing :)

The inpainting has definitely improved, a while ago it was impossible to get it to properly match the style of the rest of the image. You could always see where the original was altered. Now it blends much better with the rest of the image.
And inpainting of non-generated images is a recent thing, before that you could only alter images that Midjourney originally created.

Toxic Mutant Ninja Turtles

The inpainting has improved a lot since then. Recently they introduced an external editor that allows you to do more accurate inpainting and even retexturing.
For example, taking one of the images here.

With retexturing I can write: A 1900s photograph, of sailor moon and politicians and a xenomorph, in congress
And have it transformed while keeping the original characters:

There’s also the option to repaint:

And to expand the image:

But things it doesn’t do well is accurate stuff, like flags, characters, that kind of thing. It likes to hallucinate a little so, for example, you won’t get a perfect flag. And even a sailor moon will often look a bit off-brand.

Let your presentation match your identity :)

Prompt: 2000s TV talk show, versus, an orc and an elf sitting across from each other, in a tv talk show studio, the orc is wearing crude metal armor, the elf is wearing flowing white robes, the headline “Opposing views” is writting in bold font at the bottom of the screen, the subtitle “Hobbits, tasty or not?” is written in the bottom left, HD-TV screen shot --ar 91:51 --v 6.1

I actually managed to get the proper subtitle, but I liked this one better. Like Gollum managed to sneak in a quick edit before airing :)

I’m so devastated to see this happening and am still processing it myself.
Please take care. Best of luck and lots of love from across the pond to all of you.

For a moment there I thought I was looking at a !sciencememes@lemmy.world post and was taking this seriously :)