You're a planner in a branding agency but all the creatives are busy, and the client needs some mockups.

I know this isn't really your job
But all the creatives are busy and the client wants some mockups of how the new product shoot will look
He wants to emulate the style of Batman
You know, dark and gritty
The christopher nolan one, not the new one
Anyway, can you try and figure out the style we need?
We need a detailed description to go in the brief
Thanks

Prompt reverse engineering is a powerful tool for anyone looking to create images that match a particular style. Midjourney's new feature, "Describe," makes it easy to take an existing image and reverse engineer the prompt for it. This is especially helpful for those who are not familiar with art styles or don't know how to replicate an image.

Describe works by allowing users to type in "describe" and then upload an existing picture. It then generates four examples of what the prompt might be for that picture. This can be used as a starting point for creating a new image. It also provides keywords, such as those associated with iconic Hollywood stars or memes, to guide the user's creative process.

For example, if a user wanted to replicate a Batman movie image, they could use the Describe feature. This would generate four prompts based off the style of the original image. The user could then choose one and hit submit to generate a new image based on that style. The feature is also capable of understanding different art styles and provides similar images for reference.

The Describe feature is also extremely helpful for understanding art styles. It can provide a user with insight into what aspect ratio, color scheme, and other elements were used in the original image. This can be used to create new images that capture the same style, even if the original image is not used.

Clip Interrogator works in a similar way, but takes a longer time to run. Describe is much faster and provides more useful results. This makes it much easier for users to find the perfect image for their project.

In conclusion, prompt reverse engineering is an incredibly useful tool for anyone looking to create images that match a particular style. Midjourney's Describe feature makes it easy to take an existing image and reverse engineer the prompt for it. This feature is also useful for understanding art styles and provides helpful suggestions for creating new images. With Describe, users can quickly and easily create images that capture the same style as their original source.

Prompt Reverse-Engineering

0:01 Hey, I'm going to teach you about prompt reverse engineering. So this is a new feature in MidGeni called describe. The way it works is you just type in describe and then you can you know, an existing picture and hit enter.
0:18 And it's going to reverse engineer what the prompt is for that picture. So this isn't actually an AI generated image, right?
0:24 Like you can see here, I uploaded a picture of Batman movie. And it gives you four examples of what it might be.
0:31 So this is really useful because one is you can, you know, kind of click one of these and say, okay, I like this one.
0:39 I think the sounds the most similar. So I'm just going to hit submit and now it's kind of generating a new image based off that you know, that style.
0:52 So that's really useful. Like if you have an image that you kind of want to replicate. And, or at least as a good starting point, you can use that.
0:59 But I think where it's mostly sort of understanding what are the different art styles or similar art styles being used.
1:05 So you can see here, it's looking to David LaChapelle or Cara Cionnet. Or Dusan Jukerich you know, it's kind of showing the aspect ratio.
1:15 It's giving you different ideas, like in the style of snapshots of iconic Hollywood stars. Burned chart, right? So these are different keywords.
1:23 Memes, you can basically replicate you know, when you're generating new images. So it doesn't have to be of Batman. You could just take this and you could change it and say instead of, you know, You can make these pictures more videos instead of happening and you can see the you know, how you can
1:38 put in. Imagine you could say a spider man and put that in. Now you can get a sense of you know, what would the.
1:51 A spider man look like if he was more dark and gritty, right? And you can also, if this art styles, you don't really know about or like, you know, what types of kind of illustrations being used.
2:04 You know, you can, you can. And you can use this, which is, I think, pretty helpful. So so that's Majorini described as also the clip interrogator, which does something similar.
2:16 It takes a long time to run. But it's basically, I think, using the same sort of thing underneath. Mid-journey's much faster, it's much more useful, and it gives you f- forward it from examples.
2:27 So I highly recommend you do that, but I just wanted to kind of let you know that that's possible. So here we go.
2:31 Let's see what this Spiderman looks like. It's shaping up now, which I think is pretty cool. You know, so actually while we're doing this, this is a replication, right?
2:44 So we just took the prompt here, don't not raise as well papers inside of Duke Rich, which is the third one that we had here.
2:51 And you know, we have like a. Very similar style, right? So it's, it's replicated it, even though it's, you know, it's not the same image, right?
3:00 Like it's kind of learning from the style. So this is really, really powerful. It's really good. So it's a great way to, you know, take images that you like, especially if you don't know much about art and kind of find similar images and kind of give you a really good sign point.
3:15 So here you go. This is Christaman Ellen's Spiderman. Looks really cool. Alright, hopefully that was useful.

What is the benefit of using the Describe feature in Midjourney versus CLIPInterrogator?

Does the image need to be AI generated for Midjourney Describe to work?

How can the Describe feature be used to generate more creative images?

I don't know about you, but I didn't go to art school, so when I look at an image I don't know how to describe it in the proper terms.

Vexpower Learning Certificate of Completion

Congratulations Carina Obster 💪