Art

AI Art via GANs

Recently, I’ve been playing with rendering art via AI and am really enjoying the detail I’m getting while melding really disparate ideas.When I first encountered these machine-learning sourced image generators, I’d be able to render fun but very crude outputs. Now it’s an entirely different with the latest round of situation and this shit looks real good.

GANs achieve this realism by using a generator, which learns and produces the target output, alongside a discriminator, which learns to distinguish true data from the output of the generator. THEY ARE ADVERSARIES!

Anyway, enough talk, look at these badass image sets I was able to generate.

I started small — these are my very first midjourney renders. Aww so typically rainbow.

I moved on from “rainbow rock” to scenes in shapes. Now it’s rainbow junk in orbs.

After a few more fairly rainbowy renders It was then that I started thinking about what kind of art niche I could populate with this. Looking around at the many AI Artists and prompt jockeys — I noticed a trend for big bold fantasy and colorful chaos. I mean, why not, given the chance to have every color and so many textures come together at once in a pleasing or engrossing image?

I began to pull from parts of me that are generally unexpressed in my day to day — and also not pupular to combine. South Asian/Indian themes with more traditional elements with magic stirred in seemed like a good idea. I wasn’t quite ready to let go of rainbows.

All the while, the midjourney devs were pushing out addittional versions and options, letting us lean the output of the prompt in amazing ways. I got better at getting more consistent results with portraits and I started expanding on south asian themes.

Unbelievable. I could never produce the detail in some of these myself — alternatively, it was more of a collaboration because many of these “ideas” turned out differently than I had imagined. It was unbelievably fun to explore so many styles and ideas in rapid succession.

A whole month has passed by and I have learned a few things, and I’m still enjoying myself. Dreaming while awake – what a damn cool future man. Today I spent time dreaming up cloudy scenes.

It is easy to generate fantasy landscapes that showcase beautiful light:

I’m having loads of fun, you can try it too! https://www.midjourney.com/home/

Previous Post Next Post

You Might Also Like

1 Comment

  • Reply AI Art Update – Unbroken Horse December 14, 2023 at 10:52 am

    […] Things have been wild in the AI generative community over the last few months. All the models are developing at breakneck speeds and we are achieving some really amazing generated content in seconds. I am still in awe as I witness the changes happening in the image generation subsection of AI. It’s also just so much fun to just throw prompts in and parse through and refine results. You can see the clear difference between the generated work I am able to prompt today, versus just a few months ago. […]

  • Leave a Reply

    This site uses Akismet to reduce spam. Learn how your comment data is processed.