Few recommendations on enhancing the standard of your Midjourney photographs
In this temporary publish, I wish to share a couple of recommendations on getting extra coherent, authentic, and aesthetically pleasing leads to Midjourney.
We’ll begin with the next immediate.
a good looking Asian girl sporting a crown and a formidable costume –ar 16:9
The immediate is brief and easy, with out pointless noise, and the outcomes are already fairly good. But let’s see if we are able to get much more fascinating photographs utilizing a couple of methods.
First of all, it’s a really helpful trick to leverage archetypes as a substitute of writing detailed descriptions in prompts. Specifically, in our case, we are able to substitute “girl sporting a crown” with one phrase “queen”:
a good looking Asian queen sporting a formidable costume –ar 16:9
With this small twist, we had been in a position to get extra various and charming photographs. There are a number of the reason why this trick works for Midjourney prompts:
- It reduces the variety of phrases, and thus Midjourney may have extra space in its reminiscence to think about further particulars that you just specify in your immediate.
- It invokes particular coherent photographs that folks (and now AI algorithms) have of their reminiscence, like for instance, “cat girl”, “mad scientist”, “schoolteacher”, or “gamer.”
- Short descriptions based mostly on archetypes are extra standard for picture titles and alt textual content and thus are extra acquainted to Midjourney’s AI fashions educated on such information. For instance, it’s more likely for the picture to be titled as “a bicycle owner” than as “an individual using a bicycle.”
Read extra on Invoking archetypes within the #prompt-faqs channel of the Midjourney’s Discord server.
