Midjourney’s Crazy New Describe Feature
Helping You Write Better Prompts
Midjourney released a new ‘describe’ feature that lets you transform images-into-words.
“We think this tool will transform your liguistic-visual process both in terms of creative power and discovery.” - Midjourney team
The Importance Of Image-To-Text Descriptions
Image descriptions have important broader implications that are worth mentioning:
- Improved accessibility: Image descriptions make digital content more accessible for people with visual impairments or reading difficulties. This is done via the ALT text element for web displays.
- Enhanced searchability: Descriptions can enable better search functionality and indexing via search engines.
- Use for captions: Captions can incorporate descriptions to provide additional clarity to images.
- Detailed Prompts: Descriptions can be used to create more detailed prompts for crafting new variations. They can provide inspiration for prompt engineering.
Midjourney will generate four different descriptions based on an image you upload and makes it easy to generate new variations.
How It Works
The way it works is you simply start by writing /describe and Midjourney provides a way to upload an image.
After you upload the image, you click enter
Midjourney then returns four descriptions based on the image
The four numbers on the bottom are active remix buttons — each number matching the corresponding description. Clicking on the number will remix the image based on the new description.
You can also modify the prompt via the remix:
This is actually a cool remix version: I like it better!
This was the original prompt I used to create this sample image:
an illustration of a brain with tree roots, psychedelic art, vibrant, by Alex Grey, by Amanda Sage, by Robert Venosa, neon colors
And this is one of the prompts that Midjourney described, that I used for the remix:
An image of an abstract brain tree with roots, in the style of mark henson, luminous colors, dark symbolism, detailed anatomy, bold lines, vibrant color, psychological phenomena illustrations, chiaroscuro woodcuts
I am surprised to see how different the prompts are in comparison but are somewhat similar.
Just for fun, I uploaded a photo of the NASA astronauts via the new moon mission — and had Midjourney describe and generate a new version of AI astronauts.
NASA astronauts group pose for a photo, in the style of photorealistic portraits, dark cyan and orange, uniformly staged images, romantic depictions of historical events, celebrity portraits, hasselblad h6d-400c, non-representational — ar 117:77 — v 5
If you liked this article, throw out some Medium love… claps, comment, and be sure to follow.
You can also support my work on Medium by becoming a member using this referral link.
