
Starting with Your AI Model in Midjourney
The first step in creating a custom human image is to design your base model. You start in Midjourney. Here, you can tell the AI specifically what you need, especially if the image needs to look very real for things like product photography or client work. You design the initial look of your model piece by piece, focusing on the specific purpose you have in mind.
Refining Details with Photoshop’s Generative Fill
Once you have your starting image from Midjourney, you take it into Adobe Photoshop. Using its AI features, like the Generative Fill tool, you can add specific details or refine parts of the image. You work through it section by section.
Adding Features Piece by Piece
- You might start by adding details like freckles to the nose to make the image more unique.
- Move on to other features like the eyes. The Generative Fill tool gives you a few options each time. If you don’t like the results, you can easily regenerate for new choices until you find exactly the look you want.
- Continue adding elements like hair, the neck, and even things like earrings. Sometimes it’s effective to just tell the AI what you want (like “gold earrings”), while other times you can leave the prompt empty and let the AI suggest something based on the image.
- Adding eyebrows might seem simple, but paying attention to how they cast shadows can make the image look even more real.
Working this way allows you to have control over the final appearance, picking the best AI-generated parts to build your complete image.
Dressing and Expanding Your Scene
After focusing on the face and upper features, it’s time to get your model dressed and expand the image to include more of the body or background. You can expand the picture area in Photoshop. Depending on how complex the background or clothing needs to be, you might continue adding details bit by bit, or you can make broad requests if you just need a simple setup like a solid background or basic clothing.
For example, if you need a basic studio shot with a white shirt, you keep the instructions simple. If the first few options from the AI look good, you might stop there. If not, you can always add more detail to your prompts.
Using Your Custom Model for Branding
Once you have created your custom human image using Midjourney and Photoshop, you can actually take that finished image back into Midjourney. Use it as a reference image to generate many different pictures featuring the same model. This way, you can create a consistent visual identity with your own unique AI model for different marketing or branding needs.
Conclusion
Combining the initial creative power of Midjourney with the precise refining abilities of Photoshop’s AI gives you the ability to build detailed, custom, and realistic human images. This method allows for a high degree of control, letting you create just the right look for your projects or brand.
Do This Locally with Logen
Skip the subscription. Generate images like these on your own GPU — private, unlimited, no cloud dependency.
Midjourney vs Logen
| Feature | Midjourney | Logen |
|---|---|---|
| Cost | $10-60/mo | Free & Pro plans |
| Privacy | Cloud (images stored) | 100% local |
| Generations | Limited by plan | Unlimited |
| Internet Required | Always | No |
| Custom Models | No | Yes (any ComfyUI model) |
| GPU Required | No (cloud) | Yes (local) |
| Ease of Use | Discord bot | Desktop app |