Midjourney: a simple composition technique for capturing subtle expression
Easy and effective
Eight different techniques were discussed in the previous story about Midjourney’s Layout and Composition to show you various ways to create a stunning layout and composition for your image.
But what’s the point of learning the theories if I don’t show you some applications? Right?
In this story, I’ll show you how to use a simple composition technique to capture your character’s subtle expression.
In other words, you want to avoid crafting a prompt instructing the bot to create a character that expresses bold, intense, and too dramatic expression. You want to indirectly capture your subject’s delicate and subtle demeanor or expression.
Consider the phrase, “Tell me something without telling me.”
Before I demonstrate how to create this type of prompt. Look at the pictures below and tell me three things:
(1) Who is the main character in the photos (image grid)? (2) What else can you say about the main character? (3) What is the hidden meaning of the photos? If any
OK, take a minute or two to look at the image grid first.
Then we discuss the three things.
No cheating…
Ready?
Isn’t it simple?
The lady is the focal point, and she is the boss.
What should we say about her? (aside from being attractive). She appears confident, walks before the man, and commands the relationship.
Is there a deeper meaning here? Should we describe her as an independent, modern, and poised woman? Image 3 could also represent a sense of liberation as if she doesn’t need a relationship to be happy and content.
Here is the prompt
/imagine prompt: Jennifer and Ben walking in Paris street, street photography
Wait, are you serious? The Midjourney bot can recognize people’s names?
Yes, apparently, it can.
You can either prompt: “a lady and a guy walking” or type some names, and the bot will figure it out.
Let’s try another photo now.
This time, I’ll give you the prompt first, and you’ll have to figure out how to connect the photo to the prompt.
/imagine prompt: Ben and Jennifer walking in the street, photograph
This image grid is less obvious than the previous one. However, if you look closely, you will notice that
Image 1: Ben is slightly ahead of Jennifer
Image 2: Ben seems to be more prominent than Jennifer
Image 3: Ben’s hand is at the front, holding Jeniffier’s hand
Image 4: This is obvious. The main character is Ben.
Put the main character first in your prompt to achieve subtle expression in your image composition.
It’s as simple as that; the bot will do the rest.
There’s no need to add words or terms to your prompt to capture the subtle expression.
This technique was covered in the story “Layout and Composition.” The first subject will be the image’s focal point, with the exception of the scene/background settings.
What if you prompted the bot to place Jennifer in front of Ben?
/imagine prompt: Jennifer walking in front of Ben in Paris street, street photography
As a result, Jennifer becomes the focal point, with Ben trailing too far behind or blurred away. She is the center of attention.
There is no subtlety or nuance. The message is direct and clear.
What about the above photo?
I can’t tell who is more prominent, but I’m guessing Jennifer’s face facing the camera draws more attention than the guy’s.
And what about this image?
The lady is the center of attention.
The Midjourney bot also gives you a hint if you can’t tell the difference.
Do you see the umbrella in Image 4? The bot has labeled the focal point for us! (Just kidding)
What if, instead of two characters, you prompted a couple?
/imagine prompt: a couple walking in Paris street, photography
The bot will treat the “ couple “ as a single subject. It’s difficult to tell who is more prominent than the other.
Giving your characters more depth
You can give your characters details like the color of their shirts, the kind of clothes they wear, etc.
In the best-case scenario, just two characters. Beyond that, putting the details on the correct characters is hard.
/imagine prompt: red-shirt-David and green-shirt-Matthew walking in the street, gay couple, photography
(Note: The outcome of the prompt should be the same if you write “red-shirt David” than “red-shirt-David.” The hyphenation style is just my personal preference.)
These two guys were assigned different shirt colors. The bot did a great job and put them in the right characters. That only happens sometimes.
Most of the time, you’ll find assignment mistakes, like both guys being red or one guy being red while the shops around him turn green. That’s because the current version of Midjourney gives you little control over the colors and details. Re-rolls if the colors or other details (like facial features, body parts, etc.) were wrongly assigned.
Do put “gay couple” or “lesbian couple” (or similar terms) at the end of the prompt to show that the men or women are a couple.
If not, the intimate relationship between the characters won’t be expressed. They will become friends or strangers. Sometimes, one of the characters will be taken out of the image! (Bot, you are so mean)
Related stories
Conclusion
- Put the main character first in your prompt to capture your characters’ subtle expressions.
- Adding more words to the prompt is unnecessary and complicates things.
- Put “gay couple” or “lesbian couple” at the end of the prompt to capture the characters’ intimate relationships.
I hope you like this story!
Please give me a clap or leave a positive comment to let Medium’s algorithm know that this story is helpful and has contributed to its platform’s success.
Your support inspires me to create similar or higher-quality content in the future!
Thank you for reading and sticking with me until the end.
Never miss a story. Follow me on Medium for updates, stories, and tips about Midjourney.
Join Medium today and read all my stories:
