Once upon a time.. in the creative realm of Midjourney, we embarked on an intriguing quest: the pursuit of generating consistent characters.
We wanted to see how well Midjourney could help us create characters that would enable us to visualize the stories we want to share, perhaps even paving the way for future storyboards or marketing assets for clients.
We drew inspiration from Chase Lean for this process and infused it with our own exploration. Let's get to it.
Create a prompt for your subject's appearance and include "split into 4 different images, shot from different angles" to get multiple images of the same subject.
While it deviated from our "straight hair" prompt, we opted for the image that appeared more realistic and exhibited greater similarity among them.
Explore our previous blog post for a guide on how to use the "Vary Region" feature.
Prompt: Trevi Fountain
Apply the same process to other images. Experiment with the selection tool in "Vary Region" to tweak backgrounds and even clothing.
Prompt: Woman in snow and in winter clothes
Prompt: dressed as an employee at Silicon Valley
Click the pan right button