How It’s Made: AI Art Composite - Portrait - Vampire Queen - Indianapolis, Indiana

The Goal

I’m really fascinated with what this technology brings to the table. This will be my fourth blog now on the subject, and in this blog I want to take a more in depth look at how we actually created this image, and how we can start to use this technology to speed up the creative process.

I know I am not alone in the pursuit of what this technology can do, and what it means for the industry. I’m also aware some people really don’t see this as art.
I’ve been thinking a lot on that subject and here is what we have deduced and concluded.

Art is subjective.
Mind blowing right?

At the end of the day, if you are talking about it. It is effective art. Art lives in the gaps, in the places we don’t venture. In our emotions.
Good art forces those doors open and makes us feel something. And yes, something can be anger.

Alas, I Digress.

The point of this blog is to dig into the how, the nitty gritty so to speak. Let’s start by looking at generating an image for use in photoshop.

So let’s get this party started, from this point on I am going to assume you have read my prior articles outlining how to get started using the software or you have that knowledge. If you haven’t you can read those here.

To get started go ahead and type /imagine a vampire for your prompt. You will probably want the top plan to use this software effectively you will go through hundreds of failures to get good at communicating with the “algorithm”.
Today I am going to show each step of my discovery process, including the failures.

The only thing you will ever learn from is failure.


Before We Begin

I’d never be the type to ask anyone for a donation, but a lot of people have asked how to support my art. I love to make it, but the hard reality is I only get one hour a day to make this content in between my paying clients. If you’d love to support us, I’d be honored to have one of our prints hanging on your walls!


/imagine a vampire

/imagine a vampire

Pretty cool actually! But notice this looks more like an illustration, then an actual photograph. This is not ideal for a composite

Close, But No Cigar

So what do we do?

How do we get the image to go from boring and unoriginal computer art, to something unique and artistic?
For starters, you need to learn the commands and parameters that your chosen AI responds to.

We have been using Mid Journey, and have actually already written an in depth article about how to give the program instructions. We recommend reading that as well, if you are really interested in making art with Mid Journey.
If you want to read the article, click here.

So Now What?

We need to get more specific. And use the parameters we covered in the previous blogs linked above.

For example let’s modify our input prompt to be more specific to what we actually want. Try…

/imagine a vampire queen, ornate jewelry, haunting, beautiful, —ar 4:5

It’s About Description

I have noticed that getting any form of an anatomically correct image from Mid Journey is nearly impossible. It can be done; but you need to be targeted with your language. The AI can’t interpret like a human. So a phrase like sea lion may make it think of a lion that lives in the sea, not the actual animal. The skill of using the AI, is knowing the language. I can’t say that enough. I chose vampire queen for this blog because vampire doesn’t leave much room for interpretation. I figured it would be easier to see the effects, with less variables at play.

From here it is just a matter of experimentation and repetition to create an image you can use for your vision. It’s very important as these prompts get more complicated that you follow the prompt structure outlined below.

The Project

An Amazing Friend

Arianna has become one of my good creative buddies, and is a model ambassador for us. She helps to maintain a self community for photographers and models alike and is an amazing person to work with.

Original AI Artwork

Vampire queen, in a gorgeous generated background. I prepped this to be super eye catching. I also went ahead and desaturated the image to help with blending the two together. We are going to have to adjust either the source skin tones, or Ariannas but I have a method I will share in a later action pack to make that quick and easy in photoshop.

Swap It

Whatever your preferred method to swap faces is in photoshop this is the point where you need to get everything lined up and ready to merge. There are ways to of course make this even more shaded and full of depth, but I didn’t like the dark eyes to begin with.

Color Grade / Touch-Up

Now we clean up the image and give it our signature style, in my instance this involves cinematic color grading of course!

Photoshop Magic

This is one of those instances where a video outlining how I composite would be awesome. Unfortunately I don’t have one for you yet, but stick around and I will teach you photoshop too.
Until then, here is the final image.

Final Words

I could of spent another hour in photoshop, making this look even more real. But I made this image really to illustrate the power of this software in the hands of an artist.

I am really excited to keep deep diving into this tech, and you can always count on me to have a keen interest in these types of subjects.
Thanks again for tuning into my writing, I’m trying to keep some fresh content coming out at least a few times a week. So make sure to subscribe to our newsletter, so you don’t miss the next one!

Everything you can imagine is real.
— Pablo Picasso
Previous
Previous

How It’s Made: Poison Ivy Artificial Intelligence (A.I.) Cosplay - Astra Studios

Next
Next

How It’s Made: AI Art Composite - Portrait - Cyberpunk - Indianapolis, Indiana