Computer generated images of Midjourney
My little excursion into the world of artificial intelligence. (AI generated images)
I stumbled across an article about the Midjourney software and started to try out how good this AI (artificial intelligence) really is.
For this reason I fed the AI while sitting at home in front of my laptop with the following words: „my very tired son coming out of his bedroom holding a bis glass of beer in his hand“ (of course I made up the second part of the story).
The Midjourny software spit out the following result after a few minutes:
Since I was a bit impressed by the result, I immediately tried it again:
„man sitting in front of a sewing machine while wearing a funny hat“ (while I was thinking of indnat.com ;-)
Midjourney immediately spit out the following four suggestions, which I had a hard time deciding on, as I thought all of them were „successful“:
I decided for the last variant and told it to continue working on the last design – it generated the fourth variant in high-resolution with more details elaborated:
Next I tried:
„albert, the cooles boy in town riding on his mountainbike through the alps“
… and then I had Midjourney work out the 2nd variant – again the high resolution result can be seen:
The artificial results are worthy of attention in my opinion. If you pay attention to the details, you can see that it makes little sense in parts, but it’s not supposed to. It’s also amazing that it immediately puts a helmet on me, since I’m on my bike in the mountains.
In the last task I let the AI then fail. I let from „Albert Augustin webworker“ portray me.
It’s nice that the AI thinks I’m in my early 30s and hasn’t noticed that my hair color has changed a bit in the meantime. Also the AI probably converts „webworker“ with the socialist worker style.
I am curious where the journey continues to go and how we will see the next months / years in the public more and more AI-generated images, whether in advertising or in art.