Midjourney’s describe feature provides a powerful way to generate inspirational prompts from an uploaded image or image link. This article compiles and reorganizes the provided draft into a detailed, readable guide that covers both the basic use and the combined potential of describe with style references on the web UI.
Part One: Simple Guide to Use “/describe”
Step 1: Type the command /describe. Type the command /describe and you will be given two options: Image and Link. You can either upload an image or paste an image link that you found on the internet. Press enter or click to view image in full size.
Step 2: Four Possible Prompts. Click “enter”. Midjourney will analyze your uploaded image and suggest four prompt ideas for you. Press enter or click to view image in full size.
Step 3: Choose the preferred prompt. Users can choose from the suggested prompts or combine them to visualize new images. You can then generate images for each prompt by clicking on the relevant 1️⃣2️⃣3️⃣4️⃣ buttons underneath or select “Imagine All” to generate all four variations at a time. Press enter or click to view image in full size.
Step 4: The Results. Here are the four variations generated with the NEW “/describe”. Midjourney's /describe command is a powerful tool that allows users to upload an image, and Midjourney generates four text prompts that attempt to describe the contents of that image. Coming from discord, /describe was all you needed to type, followed by an image or an image url. Now with the web UI this feature has been hidden quite well, however once you know where it has been placed, it’s straight forward to use. It kind of makes sense where it's living but finding it is hard. Maybe that'll change as the team is making a higher effort in crafting a splendid user experience.
Before guiding you through the UI step by step it makes sense to emphasize the way /describe is intended to be used: to output prompts that are inspirational and suggestive. That’s it. It’s not a tool to recreate an uploaded image exactly (We’ll cover that part in another article).
Web UI has taken that fact to another level as you'll see below, cause if you’re comparing the output you can clearly see that the discord /describe command outputs four prompts whereas the web UI suggests building blocks for you to craft your own prompt. I really like the fact that Midjourney is taking the opportunity to rethink commands and the way they are used when moving to the web instead of just putting the same command behind a button. Alright, let's go over this.
Upload your image the same when you would want to use it as a style or character ref or as an image prompt. Once the uploaded image appears, hover over it and then click the info icon. The UI really nicely now displays individual blocks which can be used to build a prompt → Subject and Descriptor. 🙌
That’s it. Now you know how to use the describe command in the new Web UI, at least in the current version of Jun 20th 2024. To support me:📬 If you’re not subscribed yet to my newsletter Imagine Weekly, I’d be thrilled to welcome you on board!
Part Two: Combine the Power of “/describe” and “Style Reference”
The web UI enables users to leverage /describe alongside style references to craft tailored prompts. By uploading an image for reference and then using the built prompts as a base, you can guide Midjourney to incorporate your preferred aesthetics, textures, and lighting into the generated outputs. The process enhances control without sacrificing creative spontaneity.
First, upload an image that represents the style or mood you want to capture. Then, use the four generated prompts as starting points, selecting elements that align with your vision. Finally, attach a style reference or a specific artistic reference as descriptors to steer the final renderings toward your target look.
In practice, this combination helps you move from a raw concept to a cohesive result that respects both the subject matter of the image and the stylistic direction you want to pursue. The key is to treat describe as a spark that illuminates descriptive blocks-Subject, Descriptor, Style, and Mood-that you can assemble into a custom prompt.
Tips for Maximizing Describe on Web UI
- Use clear, high-quality images as references to improve the relevance of generated prompts.
- Experiment with combining multiple prompts to create hybrid concepts.
- Hover over uploaded images to access the info icon and view the building blocks of prompts (Subject and Descriptor).
- Utilize style references to guide color palettes, textures, and lighting that align with your artistic goals.
Example Workflow
- Upload a portrait image as the reference.
- Review the four prompt suggestions and pick two that emphasize lighting and mood you want.
- Combine the selected prompts with a style reference, such as “cyberpunk neon, film grain, high contrast.”
- Generate a set of images and iterate by adjusting descriptors and style terms.
NEW Midjourney Feature -- /describe
Note: The described version and UI placement refer to the Midjourney web UI as of June 2024. Interfaces may evolve with updates, so adapt the steps to your current layout while preserving the core approach of describe and style reference.