AI Imaging is here and it is Unreal. MidJourney.

MidJourney AI Imaging is here and it is Unreal. 17 Images. 

By Steve Huff

Hello to anyone reading this today, as it is something different indeed. Nope, this is not a HiFi review nor is it a camera review. Today I want to share some thoughts on Artificial Intelligence Imaging which is starting to take hold this year in 2022. When it comes to AI imaging, some services are average and some are rather amazing. I have tried all that I personally could and none have even gotten close to MidJourney for me.

Right off the bat, I ended up buying a subscription for one year to MJ as I was so impressed by it. One benefit of the yearly sub if that all you create and do is private.

With MidJourney you have to sign up on their website and also start up with Discord. Once signed up, you will use Discord to create your images. Later you can see every image you have ever made on the midjourney website under your account.

Once in you will see sections in discord to explain how to use it and there are even beginner sections to get you started. All in all it took me 30 minutes to get signed up, inside, learn the basics and have my first AI Image created.

What makes this so unique as this service has a (creepy) way of bringing forth ANY image you can imagine, as long as it is PG-13 as MidJourney does not allow any sexual images to be made, nor any with graphic violence. They want to keep it PG13 and even if you sign up for the year and get full privacy for your creations, you are still at PG13 levels ; )

To get started making images, all you have to do is go to the IMAGINE prompt within discord and type in what you want the AI to create. If you are generic in your description, the image may come back simple, basic and well, generic. If you get deep, and use prompts that the AI will recognize you can create absolute works of art using this method.

It’s not that easy though, if creating impactful art is your thing. For me, I was inside for hours on end working on just one image as you can make variations of each image or start from scratch adding in more descriptive text. Some of the images I created used long detailed descriptions, and those are my faves.

In fact, I made a series of images over one week and posted them as prints for sale. Within 3 days I sold over 80 prints. I was in disbelief but some out there felt a connection to the images I created with MidJourney. In fact, you will only be limited by your imagination and no two images will be alike or the same, even if someone were to use the same text you created for your image.

For me, once I had the final image in MidJourney I would then take it to a photo editor to enhance color, contrast or even change colors. To finish it off.

I will post fifteen images below that I created solely with MidJourney. If this sounds like fun to you, you can check out it out at MidJourney’s website and try for yourself. I have also been reading up on this tech, and it seems the artist community is worried. I share that concern but for those who draw or paint, not photographers.

Photography will never be replaced as AI will never be able to record our memories or images of our loved ones throughout life. But with the painterly quality coming from MJ, and as this tech matures, I can see some astonishing things coming from it.

Think of this as a journey of the mind, as that is what it is for me. I have been able to create images that I see in my mind, with MJ. It’s absolutely astonishing and why I splurged for the year subscription.

For this one I described a memory from my childhood for the room, and placed an elderly man into a chair watching his neighborhood and contemplating life during his last days. This looks like a painting to me (something I also added in) and this was created purely from my text, and what I had envisioned in my mind.

For this one I wanted a surreal image so I imagined a giant sunflower in a field, surrounded by other flowers and a woman admiring its beauty. 

Speaking of Beauty….

When I was a kid I always feared Tornados and this probably stemmed from watching the Wizard of Oz when I was young. For this one I wanted to create a sole house with a massive tornado trying to lift it, but failing. The way the AI created this dramatic scene blew me away. 

I made this to symbolize our furry friends living on after they leave us. I created it for someone in particular and I did so in anvil painting style. Everything from the dogs wings, type, flowers, colors and collar all came from the text I described. 

I wanted to see if I could make an image of myself and Debby walking through our woods at night, and I picked out our clothes, lighting, fireflies and a few other things and it came out just as I had imagined. 

Here I tried to make a portrait of a brown and white Jack Russell Terrier surrounded by glowing light and colorful butterflies. The description was about one paragraph in length and this is the result from AI after many variations and tweaks.

Another attempt at a surreal kind of image. This one features a woman dancing in front of her house while the world burns, In another time and another place, maybe even another dimension. The things that can be created with mid journey is incredible.

I wanted to see how close I could get to an old 1950’s photograph so I made the setting in the 1950’s, made the image B&W, and spoke about a man with hat and coat sitting in a cafe looking at the foggy air outside. This was created from nothing but the text that I wrote that described what I wanted to see. 

I am a very spiritual person (though not religious) and I made this one to symbolize a boy getting his wings before he travels into the light. This is my best selling print and it took me about 5 hours to create with MidJourney. 

I described an alley cat but with style so I gave him an old suit coat he found in a garbage can and added some rain and gloomy weather which gives this guy the “Alley Cat Blues”

Just wanted to create a lovely, dreamy, ethereal image with beauty…

This one is quirky but I love the style as I asked the AI to create an image of a skinny bald man waling with ghosts. I also said I wanted it in the style of Starry Night by Van Got and added colors, light and everything else you see here, just via words. The AI made the image. 

My 2nd best seller, the Road to Redemption…

At the Rainbow Bridge shows two dogs arriving at the same time, admiring the beauty. I had to be clear with how I wanted this to look so I described everything you see here and the AI did the rest, after several variations and add ons. 

Talk about SURREAL! I instructed the AI to create an image in the early 1900’s, with a noir feel. Cobblestone street and a GIANT arrives to walk through town. Silly? Yes! But I was still amazed that anything that you can think of can be made into a work of art. 

 

15 Comments

  1. Your images are creatively stunning (congrats), but socially terrifying. I’m not sure how I feel about them. On the one hand, AI is a direct line to feelngs and creativity, which is the same role a camera and paintbrush fill, only hyper-accelerated and robust. At the same time, the algorithmic, machine part of it opens the door to unintended consequences that we cannot imagine. Perhaps this is an evolutionary step–creating art without the physicality of toxic paint, paper, or rare earth materials that are under stress. We are going to have to be less physical in every way to survive the coming climate catastrophe–perhaps this is the next step. Either way, thanks for the post–your work is incredible and fascinating.

  2. Hi Steve! I just wanted to let you know that I have been using mid journey for creating game assets for my game voidspace for about 4 weeks now. It’s incredible. I have probably over 2,000 usable game assets already having generated about 6,000 mid journey images overall. On my Twitter you can see a video demonstration of one of the clouds that I created in mid journey being used in game. If you’re ever interested in doing a follow-up article I would love to chat!

  3. I appreciate the technology but this is a very sad day for talent, skill learning and human creativity.

  4. Steve, that’s absolutely stunning. The algorithms must be very advanced to allow this, and I love the type of rendering the developers have chosen, but kudos to your imagination and the poetic descriptions you must have provided. His Last days, Foggy day in the 50s and the last image are simply extraordinary. Thanks for sharing this!

Comments are closed.