deepdreamnights

[deepdreamnights]

hot artists don't gatekeep

I've been resource gathering for YEARS so now I am going to share my dragons hoard

Floorplanner. Design and furnish a house for you to use for having a consistent background in your comic or anything! Free, you need an account, easy to use, and you can save multiple houses.

Comparing Heights. Input the heights of characters to see what the different is between them. Great for keeping consistency. Free.

Magma. Draw online with friends in real time. Great for practice or hanging out. Free, paid plan available, account preferred.

Smithsonian Open Access. Loads of free images. Free.

SketchDaily. Lots of pose references, massive library, is set on a timer so you can practice quick figure drawing. Free.

SculptGL. A sculpting tool which I am yet to master, but you should be able to make whatever 3d object you like with it. free.

Pexels. Free stock images. And the search engine is actually pretty good at pulling up what you want.

Figurosity. Great pose references, diverse body types, lots of "how to draw" videos directly on the site, the models are 3d and you can rotate the angle, but you can't make custom poses or edit body proportions. Free, account option, paid plans available.

Line of Action. More drawing references, this one also has a focus on expressions, hands/feet, animals, landscapes. Free.

Animal Photo. You pose a 3d skull model and select an animal species, and they give you a bunch of photo references for that animal at that angle. Super handy. Free.

Height Weight Chart. You ever see an OC listed as having a certain weight but then they look Wildly different than the number suggests? Well here's a site to avoid that! It shows real people at different weights and heights to give you a better idea of what these abstract numbers all look like. Free to use.

[deepdreamnights]

The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt: a two wrestling men are standing in the fog, in the style of grindcore, radiant clusters, sharp/prickly, animated gifs, strong facial expression, photo taken with provia, light orange and dark black:: a ball with a rainbow striped coating on top, in the style of bugcore, tachisme, creased, stains/washes, light green and dark brown, cute and colorful, multilayered

#unreality #midjourney v6 #generative art #ai artwork #promptsmash #surreal #horror #public domain art #public domain #free art #auto-generated prompt

[deepdreamnights]

<s>Yeah. You know who should have it?

Disney. Universal. Adobe. Nintendo. Shutterstock. Alamy.

Every last one of them more than able to field generative AI all they want since they 'own' all that 'intellectual property'. How they got those 'rights' doesn't matter, the law says that they own every sketch, every model, every frame of every film and comic, every note and chord progression of the songbook of your culture, so it's theirs.

Those corps all get to have that power to replace artists, to reduce staff, to streamline their processes, because they "own" the "rights' to those works. But you, me, and every small studio out there shouldn't be allowed to use the tech to compete on that level.

Isn't that just the best outcome imaginable?</s>

That's what you're really saying when you fall back on the copyright argument regarding generative AI. That is the inevitable result.

Adobe, Disney and the rest of the Copyright Alliance isn't suing the publicly-accessible generative AI systems and supporting lobbying to expand copyright to cover styles for your benefit or the benefit of artists. They're doing it because they see an avenue to regulatory capture of new tech.

Because scraping of public facing published work is fair use. Any human being could study any given work of that sort and take whatever they learned from it to apply to their own creations. The existence of a mechanical apparatus in that process does not change this.

Making scraping not fair use would require moving the point at which infringement happens from the point of publication to any point in the creative process. That basically removes transformation as a fair use concept entirely, since it's no longer about if the final work is insufficiently different from the inspiration, but whether any aspect of the inspiration was involved 'without permission'.

You should also remember: there is nothing in your work that can't be traced to something one of those companies owns. It may not be a legitimate connection, but who are the courts going to believe, you or a pack of highly paid IP lawyers?

As always, I direct everyone to the Electronic Frontier Foundation's statement on the subject.

#AI discourse #saying it again because it needs to be said

[deepdreamnights]

"He creeps me out but my kids can't get enough of him."

The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt: a purple and yellow wool muppet that is wearing a blue jacket, in the style of webcam photography, made of feathers, flickr, dinopunk, deconstructive, meatpunk, salon kei:: A greenish yellow humanoid creature with glowing eyes and large hands wearing a formal suit, giving a speech at the pulpit in a church, screenshot from a video game, in the style of unknown artist.

#unreality #midjourney v6 #generative art #ai artwork #public domain art #public domain #free art #auto-generated prompt #uncanny #horror #promptsmash

[deepdreamnights]

Some quick tips for peeps using MJ that don't want it to look like everyone else using MJ.

1. The Ultra Stylish All Dress Alike

First, adjust your style setting (--s (number)). This is how much of the Midjourney 'secret sauce' is added. The lower this number, the closer to your prompt the style will be, at a cost of coherence and 'prettiness'.

Same prompt, same seed, style 500 on the left, style 25 (my preferred setting) on the right. The differences in lighting, pose, skin reflectivity, etc are apparent. You can think of it as a "de-instagram" setting.

3. Use your Variations

"A dinosaur astronaut", original gen, vary (subtle) and Vary (strong) results

If you get something you like for an initial result, the first thing I tend to do is immediately to a Vary (Strong) on it. The results are usually better than the original one because you're essentially re-running the initial prompt with the first result as an additional image prompt, reinforcing the subject.

Using subtle variations to get closer to what you want is one of the big features of MJ, and it's sorely underused.

I recommend setting to low variation mode so your normal variations are subtle ones, and putting on remix mode. Remix mode prompts you to change the prompt every time you do a variation. If you're close but something's not right, that's an easy way to go.

Changing prompts is especially useful if combined with Vary (Region). Basically, if you like everything about a pic but some select details, you can highlight an area and have the system produce new variations just changing that region. It's obvious use is fixing generation errors, but by changing the prompt, you can get results that the robot can't imagine as a solo prompt.

3. Get Weird With It

--weird is a highly unused setting. It's also very powerful, while it goes up to 1000, I tend to not go above 1-5.

A cartoon penguin made of knives at 0 weird, 1 weird, and 50 weird:

Near as I can tell, weird restricts the 'does this make sense' checks, allowing for more out-there results, both in terms of style and subject matter. At the very low levels (1-5) it increases prompt adherence, at higher levels it reduces it substantially.

4. Prompt Big and Blend your Concepts

MJ deals with short prompts by filling in its own leanings where there's gaps, so short prompts look the most midjourney-esq while longer ones (especially when combined with --weird or a low --s value) fight back against it more.

When prompting for art styles, prompting for multiple art styles/artists at once produces weird hybrids. Prompting an artist with the wrong medium (A painting by a sculptor, a drawing by a cinematographer, etc) also can produce new, strange results.

(Lisa Frank/H.R. Giger style mashup using :: technique (below))

But the real trick is in "prompt smashing" or multi-prompting. Basically, midjourney uses :: to split prompts. Intended function is to allow you to add individual weights to each section, if you want something strongly emphasized.

But in practice, it blends the concepts of the two prompts to create a new, third thing.

As above with "an illustration of an armored dinosaur in the woods, in the style of vibrant comics, toycore, ps1 graphics, national geographic photo, majestic elephants, exotic, action painter" then "daft punk & particle party daft punk world tour, in the style of romina ressia, polished craftsmanship, minimalistic metal sculptures, ultra hd, mark seliger, installation-based, quadratura" and then the two prompts run together separated by ::.

Each one was also iterated to produce a better result than the initial gen.

5. Just edit the darn thing.

Learning to edit your works outside of the AI system will always improve your work beyond what the machine itself can do. Whether it's just the simple matter of doing color adjustment/correction:

Or more heavy compositing and re-editing combined with other techniques:

Editing and compositing your gens is always superior to just posting them raw, even if its just a little cropping to get the figure slightly off-center or compiling 20ish individual gens into a single comic-style battle scene before recoloring it from scratch.

But at the very least, before you post, go over the image and make sure there's nothing glaring.

I don't think that making or using AI art/image generation is morally wrong, as you guys know, but I have to admit that slotting a wibbly-lined low effort image selected from the first result set from Midjourney in your content is incredibly tacky. At least select something that looks good. Maybe something without the boring AI "sheen" look either.

You see this a lot in clickbaity content like web spam and youtube shorts attempting to algorithm game.

#AI assisted art #ai art tutorial #the uncanny valley #midjourney #ai edits

[deepdreamnights]

Facing the Past

The markers of the old world yet resist the vines of the new.

Prompt: a large man dressed as a bigfoot walks through the woods, in the style of photography installations, subdued pointillism, contemporary animal sculptures, moody and atmospheric, angura kei, made of vines, made of rubber:: homeroom horror cd art download, in the style of cyclorama, otherworldly fun, d&d, intricate use of hatching, chromatic aberration, immersive environments, dinocore

#unreality #midjourney v6 #generative art #ai artwork #public domain art #public domain #free art #auto-generated prompt #promptsmash #landscape #ai landscapes

[deepdreamnights]

AI detection is the big area for scammery now, both large and small.

On the small end, you see stuff like the above AO3 bot scam. There's no real tech behind it, and the concern that it'd be used to scrape more material is rather silly. A webspider would do the job far more efficiently. The site it's taking you to is more likely a mass of trackers and malware scripts.

But pay attention to the wording used:

"call out all AI using cheaters" is very telling language. For a scam to work, the victim has to want something, and they've identified the very common want of an opportunity to unleash cruelty on a 'deserving' target without fear of retaliation.

I have anon asks turned off for a reason.

Here's the truth of it: if there's an AI written fic on AO3 that hasn't been edited enough to count as the human author's work, then it's going to suck, and that's enough reason to not recommend it. If the answer to "why do I want to know if this is AI?" is "so I can attack the person if it is" then you need to reconsider your motivations.

The grand irony is that the fervently anti-AI crowd isn't fighting the phenomenon, they're 100% a part of it.

Because AI-hype and anti-AI paranoia are two sides of the same coin.

A realistic understanding of the situation with functional knowledge of how the technology actually works greatly reduces both the terror generative AI produces, at the same time it deflates the kind of deceptive promises that pump the company stock.

Programs that promise to detect AI and those that promise to poison datasets don't need to actually work to look promising enough to draw investor cash or a buyout, and are the new wild west of AI snakeoil.

Glaze and Nightshade are ineffective for multiple reasons and Artshield only works because AI scrapers don't want to scrape their own gens to reduce recursive dataset issues, and it uses stable diffusion to put a stable diffusion watermark on the picture (ironically making it so that it triggers AI detectors as AI). In all cases, traditional watermarks and/or signatures are more effective at dissuading random scraping.

So images can be detected as AI... provided they haven't been modified or edited enough that the AI-generator's own watermark is erased. If that hasn't happened, you don't need fancy detectors, there's free utilities to do that (I'm sure). If it has been edited, then you're back to reading tea leaves.

But there's no marking text in the same way. Some efforts have been made to make the programs embed word-patterns that can be used as a watermark, but a few editing passes or even a run through a second 'reword this better' AI and that's gone.

But there's always someone out there to sell a detector if someone wants to find something bad enough. Drugs, bombs, water, or "cheaters" -

- so long as you don't mind buying a dowsing rod.

Quick PSA, if you get one of those "Work scanned, AI use detected" comments on AO3, just mark them as spam.

Some moron apparently built a bot to annoy or prank hundreds of authors.

There is no scanning process, your work doesn't actually resemble AI writing, it's all bullshit. Mark the comment as spam (on AO3, not the email notification you got about the comment!) and don't let it get to you.

[deepdreamnights]

i already mentioned in the past that a friend of mine trained an AI on my drawings so that it could replicate my artstyle, a fun thing is that it fed it a bunch of drawings i made of myself so most of the things it creates look like self portraits i could have done

other just look like random cool guys i could have come up with

(i fucking love the guy with short hair and lipstick, he looks fucking sick, i am 100% using that design)

first of all, i cannot begin to descrive how mindfucky and trippy is to see drawing that *could have been done by me* but werent. every time i see them i have to catch my self for a few seconds wondering "wait, did i do these and forgot????"

second, because i didnt do these they are not worn on my eyes like a drawing i would have made, these are fresh images to my brain. i dont have a memory of having done every line and curve and having observed it a million times as i was doing them, thus the illusion of life is a lot more stronger for me.

so with all this is mind, this little excercise is allowing me to do something i always craved for but that i could never get. to see my own art from an outsider's perspective. to be able to appreciate my work from the point of view of someone else. how does my drawings look to others. i can do that now with these.

and let me tell you, not to sound too conceited but, damn, i really like what i see. i can now see what everyone else sees in my artstyle, its fun! its dinamic! its really expressive in a way i never noticed! and yes, the noses! oh my god the noses! and the way the lips curl and tighten.

on top of that, as i said, a lot of the overfitting in the model is directed towards self portraits that i made of myself, so the AI will tend to use many of the faces and features i use to depict myself. now you have to understand i identify quite strongly with my own drawings. i said often that i see more of my self in my drawings than i do in a mirror.

with this in mind its also super trippy that my friend basically has a fip generator.

every time i see things like this i cant help but go "OMG, THATS ME! THAT IS ME!"

(this final piece was actually done by me, lol)

so yeah, this has been all very illuminating

[deepdreamnights]

The Machine Bison

The only image from the scanning system that survived the assault.

The image(s) above in this post have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt and process under the fold

Prompt: a giant mechanical fortress shaped like a bison, retrofuture elements, the 1980s concept of the future, sci-fi landscape

Using the following style transfer piece as a style reference:

#unreality #midjourney v6 #generative art #ai artwork #public domain art #public domain #free art

[deepdreamnights]

The Shores of the Cocytus

It will not be your favorite vacation, but it's not like you'll remember it.

The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt: a large, scary city in dark colours, in the style of light crimson and azure, epic landscapes, burned/charred, piles/stacks, chad knight, high detail, mystical realms:: large tropical island isolated on the clear blue water stock photo, in the style of aerial view, sabattier filter, afro-caribbean influence, high quality photo, creative commons attribution, fujifilm x-t4, psychedelic surrealism

#unreality #midjourney v6 #generative art #ai artwork #public domain art #public domain #free art #auto-generated prompt #landscape #ai landscapes #hades #hell #hellscape

[deepdreamnights]

Public Domain AI Image Grab-Bag 12

They're cute, they're many, they're largely uncanny...

it's the Return of the Revenge of AI Ladies!

Donations/Tips are accepted via paypal.

These images are free to use, all generated within Midjourney. These unedited images were not subject to sufficient interference by humans to meet the minimal expression standard, and are, as such, public domain.

[deepdreamnights]

Grandma Knows How to Handle the Daily Grind

Do not try this at home. Always use appropriate safety equipment.

The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt: marilyn monroe appears on street with skate board and moves in a white dress, in the style of stylish costume design, loose gestures, movie still, white and gray, heatwave, flowing fabrics, uniformly staged images :: old woman in a with glasses and glasses, in the style of curves, light navy and dark magenta, webcam, cabincore, manapunk, playful poses, brown and blue

#unreality #midjourney v6 #generative art #ai artwork #skateboarding #grandmas #public domain art #public domain #free art #auto-generated prompt #ai girl

[deepdreamnights]

!7th Century French Arcade

The opulence of French arcades under the Sun King represents one of the greatest historical gaps between rich and poor gamers, and was one of several flashpoints leading to the revolution.

The image(s) above in this post have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt: photograph of a 17th century French arcade, rococo and wood panel, steampunk video arcade, many small details, filgree, professional lighting

#unreality #midjourney v6 #generative art #ai artwork #arcade #rococo #public domain art #public domain #free art #auto-generated prompt

[deepdreamnights]

Screaming in Pain(t)

Make some noise.

The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt: a pair of cartoon characters with red, blue and yellow faces, in the style of dracopunk, action-packed scenes, canon ae-1, dinocore, strong lines, mythological references, hd:: colorful paint splash from the mixing of colors, in the style of mike campau, graphic fluidity, black paintings, victor moscoso, photorealistic pastiche, paul barson, richly colored

#unreality #midjourney v6 #generative art #ai artwork #public domain art #public domain #free art #auto-generated prompt #dripping paint #paint splatter #promptsmash

[deepdreamnights]

There's a nuance to the Amazon AI checkout story that gets missed.

Because AI-assisted checkouts on its own isn't a bad thing:

This was a big story in 2022, about a bread-checkout system in Japan that turned out to be applicable in checking for cancer cells in sample slides.

But that bonus anti-cancer discovery isn't the subject here, the actual bread-checkout system is. That checkout system worked, because it wasn't designed with the intent of making the checkout cashier obsolete, rather, it was there to solve a real problem: it's hard to tell pastry apart at a glance, and the customers didn't like their bread with a plastic-wrapping and they didn't like the cashiers handling the bread to count loaves.

So they trained the system intentionally, under controlled circumstances, before testing and launching the tech. The robot does what it's good at, and it doesn't need to be omniscient because it's a tool, not a replacement worker.

Amazon, however, wanted to offload its training not just on an underpaid overseas staff, but on the customers themselves. And they wanted it out NOW so they could brag to shareholders about this new tech before the tech even worked. And they wanted it to replace a person, but not just the cashier. There were dreams of a world where you can't shoplift because you'd get billed anyway dancing in the investor's heads.

Only, it's one thing to make a robot that helps cooperative humans count bread, and it's another to try and make one that can thwart the ingenuity of hungry people.

The foreign workers performing the checkouts are actually supposed to be training the models. A lot of reports gloss over this in an effort to present the efforts as an outsourcing Mechanical Turk but that's really a side-effect. These models all work on datasets, and the only place you get a dataset of "this visual/sensor input=this purchase" is if someone is cataloging a dataset correlating the two...

Which Amazon could have done by simply putting the sensor system in place and correlating the purchase data from the cashiers with the sensor tracking of the customer. Just do that for as long as you need to build the dataset and test it by having it predict and compare in the background until you reach your preferred ratio. If it fails, you have a ton of market research data as a consolation prize.

But that could take months or years and you don't get to pump your stock until it works, and you don't get to outsource your cashiers while pretending you've made Westworld work.

This way, even though Amazon takes a little bit of a PR bloody nose, they still have the benefit of any stock increase this already produced, the shareholders got their dividends.

Which I suppose is a lot of words to say:

[deepdreamnights]

The Battle of Planet 'Za

As commemorated by Chef Zorval'Zekk. The figurines are entirely fondant free.

The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt: a handmade pizza that has small figurines on it, in the style of post-world war ii art, animated gifs, war scenes, realistic human figures, light black and violet, 1900–1917, dynamic action:: computer space adventures | computer space adventures lp, in the style of adventure themed, vhs, award-winning

#unreality #midjourney v6 #generative art #ai artwork #public domain art #public domain #free art #auto-generated prompt #surreal #pizza

[deepdreamnights]

Angel of Harmthem

You can almost hear the boss music.

The image(s) above in this post were made using an autogenerated prompt and/or have not been modified/iterated extensively. As such, they do not meet the minimum expression threshold, and are in the public domain. Prompt under the fold.

Prompt: a large rock in the background, in the style of colorful fantasy realism, dragoncore, light green and light amber, varied brushwork techniques, contrasting lights and darks, fantastic imagery, goosepunk:: Full body digital illustration of a female demon wearing a metallic black and white costume, with red eyes, sharp claws on her hands, wings, and chains in the style of Yoji Shinkawa and Final Fantasy, against a dark background, as a fantasy character design sheet in full color.

#unreality #midjourney v6 #generative art #ai artwork #public domain art #public domain #free art #auto-generated prompt #ai girl #demoness #succubus #villainess #character design #promptsmash