#ai art workflow

[onepawproductions]

Timelapse of the latest Dramione Lady of the Lake by Colubrina art, for the podfic!

Im only just starting to embark on my "timelapse learning journey", so this timelapse starts in the middle because I saved the first hour wrong and it got lost into the ether :/ I expect I'll get better at the timelapse thing as it goes on. The tentative plan is to learn how to do timelapses on my tablet, and then use Remote Desktop to access my PC so that the timelapse will show both the hand-painting in Sketchbook Pro, and the image generation in A111

After that step, I shall endeavor to learn how to *gasp*, edit video! ::dies slightly inside::, then make a narration track, before finally ::dies completely and then raises again as an ai tool art zombie:: make a YouTube channel on how to do Ai Composition Art. Ai Comps for the win!

[the-storm-chaser]

//I don't want to add this rule, but this is moreso to cover all bases. I've had to include this in all my art ToS on Vgen, so it's going here as well

-

I will unfollow/block if I catch you using AI in your writing/art/icons. My stance is that Generative AI has no place in any creative space and I have zero tolerance in anyone using it.  We are here to be creative. Using Gen AI defeats the purpose of that.

-

If you don't agree, there is an unfollow button right there. 👋

[mlearningai]

Each AI agent is a micro-intelligence,

capturing new ideas from every interaction

Merging these ���checkpoints” fuels a resilient,

creative hive mind using fewer steps and no-code workflows

The future of AI agent design is here!

[ottopilot-ai]

Anatomy of a Scene: Photobashing in ControlNet for Visual Storytelling and Image Composition

This is a cross-posting of an article I published on Civitai.

Initially, the entire purpose for me to learn generative AI via Stable Diffusion was to create reproducible, royalty-free images for stories without worrying about reputation harm or consent (turns out not everyone wants their likeness associated with fetish smut!).

In the beginning, it was me just hacking through prompting iterations with a shotgun approach, and hoping to get lucky.

I did start the Pygmalion project and the Coven story in 2023 before I got banned (deservedly) for a ToS violation on an old post. Lost all my work without a proper backup, and was too upset to work on it for a while.

I did eventually put in work on planning and doing it, if not right, better this time. Was still having some issues with things like consistent settings and clothing. I could try to train LoRas for that, but seemed like a lot of work and there's really still no guarantees. The other issue is the action-oriented images I wanted were a nightmare to prompt for in 1.5.

I have always looked at ControlNet as frankly, a bit like cheating, but I decided to go to Google University and see what people were doing with image composition. I stumbled on this very interesting video and while that's not exactly what I was looking to do, it got me thinking.

You need to download the controlnet model you want, I use softedge like in the video. It goes in extensions/sd-webui-controlnet/models.

I got a little obsessed with Lily and Jamie's apartment because so much of the first chapter takes place there. Hopefully, you will not go back and look at the images side-by-side, because you will realize none of the interior matches at all. But the layout and the spacing work - because the apartment scenes are all based on an actual apartment.

The first thing I did was look at real estate listings in the area where I wanted my fictional university set. I picked Cambridge, Massachusetts.

I didn't want that mattress in my shot, where I wanted Lily by the window during the thunderstorm. So I cropped it, keeping a 16:9 aspect ratio.

You take your reference photo and put it in txt2img Controlnet. Choose softedge control type, and generate the preview. Check other preprocessors for more or less detail. Save the preview image.

Lily/Priya isn't real, and this isn't an especially difficult pose that SD1.5 has trouble drawing. So I generated a standard portrait-oriented image of her in the teal dress, standing looking over her shoulder.

I also get the softedge frame for this image.

I opened up both black-and-white images in Photoshop and erased any details I didn't want for each. You can also draw some in if you like. I pasted Lily in front of the window and tried to eyeball the perspective to not make her like tiny or like a giant. I used her to block the lamp sconces and erased the scenery, so the AI will draw everything outside.

Take your preview and put it back in Controlnet as the source. Click Enable, change preprocessor to None and choose the downloaded model.

You can choose to interrogate the reference pic in a tagger, or just write a prompt.

Notice I photoshopped out the trees and landscape and the lamp in the corner and let the AI totally draw the outside.

This is pretty sweet, I think. But then I generated a later scene, and realized this didn't make any sense from a continuity perspective. This is supposed to be a sleepy college community, not Metropolis. So I redid this, putting BACK the trees and buildings on just the bottom window panes. The entire point was to have more consistent settings and backgrounds.

Here I am putting the trees and more modest skyline back on the generated image in Photoshop. Then i'm going to repeat the steps above to get a new softedge map.

I used a much more detailed preprocessor this time.

Now here is a more modest, college town skyline. I believe with this one I used img2img on the "city skyline" image.

[myothertardisisonthemun]

Kids these days are gonna end up making their weird images with AI, instead of coming up with the idea themselves, and Photoshopping it like God intended.

If Adobe hadn't made itself so expensive, and the free equivalents weren't comparably mediocre, we wouldn't be quite in this position.

[katdbee]

Workflow for generating 25 images of a character concept using Automatic1111 and Control Net image diffusion method with txt2img;

Enable Control Net , Low VRAM, and Preview checkboxes on.

Select the Open Pose setting and choose the openpose_hand preprocessor. Feed it a good clean source image such as this render of figure I made in Design Doll. Click the explodey button to preprocess the image, and you'll get a spooky rave skeleton like this.

Low VRAM user (me, I am low VRAM) tip: Save that preprocessed image and then replace the source image with it. Change the preprocessor to none, and it saves a bit of time.

Lower the steps from 20 if you like. Choose the DPM++SDE Karras sampler if you like.

Choose X/Y/Z plot from the script drop down and pick the settings you like for the character chart about to be generated. in the top one posted I used X Nothing Y CFG scale 3-7 Z Clipskip 1,2,3,7,12

Thanks for reading.

[dominaexmachina]

Heroes of whining

It’s amazing to watch how AI suddenly turned ordinary artists into holy sufferers.

“You’re doing so well, still drawing… despite AI…”

Well. Maybe it’s time to create similar appreciation posts for engineers, doctors, teachers, accountants, designers, biologists, lawyers, logistics specialists, marketers — all of whom learn new tools, frameworks, new medicines, protocols, research reports, laws, requirements, and interfaces every year.

You’re doing great too. But no one tells you:

“Despite spreadsheets looming over us, you’re doing amazing, sweetie.”

Here’s the reality: If you use AI as a tool — not a fetish of fear — you save time on drafts, base structure, and composition. And you can invest that time into quality, detail, and your own artistic magic.

Let’s break it down:

How does an artist work?

(One complex art piece — from idea to completion)

1. Concept. What’s the idea? Mood? Scene? Who’s in it, where’s the light coming from?

2. Sketches. 5–15 mini roughs or thumbnails. On paper or digital. A few hours to a whole day.

3. Composition. Black-and-white value layout. Masses, rhythm, focus, readability.

4. References. Photos, 3D base poses, perspective, textures, clothing — often hours of search.

5. Linework. Base outlines, underpainting, color blocking. Only then the “real” part starts.

6. Light, color, depth. Building atmosphere, spatial layers, focal contrast.

7. Character detailing. Eyes, hands, fabrics, expressions, skin, drapery, texture.

8. Background. Architecture, environment, plants, particles, reflections.

9. Micro-details. Dust in the light, droplets, stitching, jewelry, tiny surface effects.

10. Final polish. Merging, cleanup, final light tweaks, subtle accents.

🕓 Total: 30–50 hours. And still — compromises. Artists get exhausted by step 3. Backgrounds get flatter. Composition gets simplified. Details trimmed down. Because time and energy are limited.

Now imagine:

Steps 1–5 are done by AI. The artist gives direction, prompts, edits, references. But now they have all 30–50 hours for steps 6–10. And they’re not tired.

What’s the result?

— All effort goes into depth and quality, not grinding through sketchwork — More time for variations, storytelling, wild lighting decisions — Space to polish every surface: metal, skin, reflections, mood — Room for narrative in every corner of the scene

This isn’t laziness. It’s strategic effort management. The AI-assisted artist doesn’t skip the hard part — they just skip the routine part.

And yes — they both spent 40 hours. But one built from zero. The other built magic on top of a structure.

Now tell me: Which piece will be deeper? More detailed? More surprising? Which one will make people zoom in, explore, return to it again and again?

[venezart]

ComfyUI | Prompt Share | Nomad Vehicles | Cyberpunk | Illustration | PGI

Today, we are creating innovative Cyberpunk Nomad vehicles. To help you get started quickly, we’ve included the workflow.json file below. Enjoy, and don’t forget to comment, like, or subscribe to receive updates in your email whenever new ComfyUI prompts and workflows are available. Positive prompt Futuristic post-apocalyptic vehicle in a desert or urban wasteland, inspired by Mad Max and…

[onepawproductions]

[evartology]

Pro AI Design Workflows:

A Beginner's Guide

🟣

Your #evARTology creative fuel can for creating stunning visuals with AI, inspired by the pros

[mlearningai]

AI automation vs AI agents: n8n + Claude Code Quick Start Build Your First AI Workflow Automation. A Simple and Easy Beginner's Guide

[lwoorl]

"I'm against AI art—"

Oh yeah sure, most of the companies developing these models do it in ways that are very exploitative of the people they hire for classifying data, and there's also the issue of the training dataset itself. Of course, it's always possible for someone to run a model locally which is much more ethical, which is a thing many AI artists do already, but it's still reasonable to be wary of—

"—because it has no soul and isn't real art and I can always tell when it's AI because I can magically perceive the soul of an art piece and also the value of art is always proportional to the effort it goes into it and coming up with concept and ideas is meaningless!!!"

Oh fuck off you entitled gatekeeper.

[thecultofproductivity]

X.x y feel like all art I tuch bad.

Making animations feeling fake af because I didn’t make the rig from the ground up like it’s 2013.

Also the animations are kind of just actually disfocused like sh*t bc perfecting the workflow is genuinely the important part even just for this project f my development as an artist.

Okay cool I feel slightly better. Art art art

[lightarts]

Grow a home with me! Let's make hemp homes from farm to tables everywhere. I'm working to build this as a production that's participatory and easy to engage for DIY creators and builders everywhere Vote us up now at HempHome = https://artizen.fund/index/p/hemphome

[robolovely]

Here are a few of my collected failures to make Midjourney images more raunchy/explicit via the use of Subtle Variations.

First was these two nice ladies:

Honestly the base image was great, I think the intent here was to increase bust size though, which didn't get too far (some images from the attempts).

Great hats though! I'm very fond of these two. Maybe I'll use the new Midjourney website to bring them back. They were just too classy for me to debauch.

---

Next was this nice reclining woman in a painted style:

Her nose is so cute! I'm also quite fond of her. Anyway she was quite resistant to my Subtle Variation attacks on her character. See below:

I achieved some minor big boobe and tasteful cleavage but really nothing to write home about compared to my later (and earlier) works.

---

Another one to highlight is this princess in a puffy dress against the night sky:

This one was all a bit Eurotrash. I /think/ I was trying to depoof her dress, trying to make her more slinky before evolving her further but she really just didn't want to give up the puff. I've hit the image limit on this post so I won't show you the results but they're basically exactly the same image with maybe a liiiittle more cleavage. Complete failure!

I believe this puffy princess was my first attempt at using the Vary (Region) function as well, which I would use with great skill later on!

Follow for more lovely Midjourney lady art and a little bit of the process behind it. I've since managed to create truly NSFW images with Midjourney so it's come a long way since these attempts.

[plotpulse]

Unleash Your Creativity and Earn Rewards with Weekshot!