Using the Midjourney Editor to Change Character Expressions

For one of my recent projects, I had to generate several illustrated character images. Each character needed both a sad and happy expression. While I could have used Midjourney’s character reference feature to generate consistent character images, I decided to try something different. This time I used Midjourney’s new editor to generate additional expressions. Using the Midjourney editor meant I kept the same pose but just changed the expression. It worked better for this particular activity, where I was focused on portraying a transformation from sad to happy in the characters.

Image generation and editing process

Here is the basic process I used:

1. Generate the initial character image with one expression or emotion.
2. In the editor, use the eraser brush to remove the eyes, mouth, and eyebrows if needed to change the expression.
3. Adjust the message to describe the new expression or emotion you want.
4. Choose the best-edited image.
5. Adjust and edit if necessary. Sometimes you need multiple edits and commands.

In the online activity, the characters fade from sad to happy expressions. Because most features remain the same, this makes the resulting transition very smooth.

Use the Midjourney editor

In the web interface for Midjourney, you can access the new editor on the right when viewing a single image.

In the basic editor, the eraser brush is active by default. You can brush the parts of the image you want to change. In this case, I want to change the eyes to be open and the mouth to smile. These are the sections I delete from the image (creating a very freaky image!).

I also changed the command to describe what I want to change about the expression.

Original request:

Vector illustration, a young girl feeling sad, head and shoulders, looking down

Revised assignment:

Vector illustration, a young girl feeling happy, head and shoulders, eyes open, slight smile

Problems with editing

Actually, when I originally tried this, I didn’t specifically say that her eyes should be open. This meant I got the smile well, but all the images kept her eyes closed. Even with “eyes open” in the prompt, one of the four choices had her eyes closed. In the image below, the bottom row shows four versions of the source image that I generated to change the aspect ratio from a previous generation. The middle row is my first edit, without specifying that her eyes should be open. The top row is the eyes-open version.

As you can also see in the top row, the eyes are not always proportional. This has been a problem in many of my generations of revisions; the eyes were not consistent or did not completely match the original image.

This worked well enough when I only needed two expressions for each character. I think if I needed 4 or 5 expressions for each character, it would have taken more time—possibly so much time that I would have been better off editing at least some of the images in another tool besides Midjourney.

Illustrated characters work better than photorealistic characters

This approach with the Midjourney editor works better with illustrated characters than with photorealistic characters. For this project, I used illustrated characters with a fairly simple style, making editing much easier. With photorealistic characters, I find it really hard to make the person look the same. This is true even when I use the -ref function to create a consistent character based on a reference.

For example, here is a photorealistic character with a frustrated expression that I generated previously, plus four edited versions with slight smiles. Some of these might be okay as a match if you didn’t look too closely at them, but none of them are great. They look like contestants in a similar competition rather than the original character.

Midjourney for images for e-learning

While not perfect and sometimes requiring multiple attempts, Midjourney is already a viable tool for creating images for eLearning. I find that in most situations I can generate images in Midjourney faster than I can get stock images.

This is especially true for character images and when I need multiple specific images in the same style. Getting multiple images in the same illustration style has always been a problem in stock images; this is one of the reasons why I tended to use photographs instead of illustrations. But Midjourney changed that process for me.

Plus, if I spend less time searching stock image sites, I can spend more time on better writing or other activities to improve the quality of my work.