• home
  • FaceBook
    • FaceBook Hacks & tutorials
    • FaceBook Scripts - Facebook Extensions
    • Paid Facebook Scripts
  • Social Media
    • Social Applications
    • Social Extensions
    • Tech & SocialMedia
  • Websites
    • Websites Development
    • Websites Security
  • IT Highlights
    • Vulnerabilities & Exploits
    • Documentations
    • Personal Security
  • Khalil
    • contact
    • Order Web Application Scanning
    • report a bug
    • Terms of Use
  • C&M
  • Videos

Khalil Shreateh specializes in cybersecurity, particularly as a "white hat" hacker. He focuses on identifying and reporting security vulnerabilities in software and online platforms, with notable expertise in web application security. His most prominent work includes discovering a critical flaw in Facebook's system in 2013. Additionally, he develops free social media tools and browser extensions, contributing to digital security and user accessibility.

Get Rid of Ads!


Subscribe now for only $3 a month and enjoy an ad-free experience.

Contact us at khalil@khalil-shreateh.com

Mastering Midjourney's inpainting

Details
Written by: khalil shreateh
Category: Technology
Hits: 1068

Users testing the feature have already discovered incredible capabilities. Bookmark for future reference.

Midjourneys inpainting

 
Here are 10 amazing discoveries so far:
 
- Creating consistent characters with inpainting
 
Midjourney Consistent Characters with Inpainting:
 
1. Create a 'Reference Character' upscale and copy the image URL address Prompt Below each image.

Actor Casting portfolio, unusual french woman, with freckles, imperfect skin, model pose, red lipstick, flirty attitude, minimalist, grey background, kodak Porta 800 film, analog --ar 4:5
 

Actor Casting portfolio, Old man, imperfect skin, bad attitude, minimalist, grey background, kodak Porta 800 film, analog
 
2. Create the scene and use the 'Vary Region' Function to paint out the face. Prompt Below each image.
 
 

cinematic closeup of an woman explorer a blockbuster sci fi movie on a barren ice planet at night in the style of Denis Villeneuve white grey teal and crimson red cinestill 50d --ar 21:9 --style raw
 


 
3. Add the 'Reference Character' URL to the start of the prompt and edit any details.
 
Tip: A couple of people have pointed out that you need to have remix mode turned on in /settings
 
- Filling in gaps and removing items
Filling in the gaps with Midjourney InPainting
Early observations This new feature gives you the control to fill in the gaps, prompt and remove items. Now let's test it out and see if this blows you away
 
Your browser does not support the video tag.
 
 
 - Inpainting with your own image
 Midjourney INPainting With your Own Image
You can follow this tutorial to generate your own image
Step 1 - Use the below prompt to generate the below image. Prompt - beautiful white clouds shaped as ship --style raw
 
 
Step 2 - Now click on Vary (Region) and mark the region where you want your image to appear. Make sure you have the link to your image., by uploading the image to Midjourney and copying the URL link from the browser. Step 3 - Enter the above URL and a Prompt along with it (Captain of the ship
 
 
Step 4 - Now you should see the InPainting image.
 
 
You can continue to upscale, zoom and Pan this image to get different variations.
 
- Applying scaffolding techniques to inpainting
 

Here's an example of a workflow that I came up with utilizing the new inpainting feature:

✅- Start with a basic sketch
✅- Prompt the desired result over the sketch, disrupting it with an image vector for composition and scale
✅- Subtle Variations until desired cohesion
✅- Strong Variation
✅- Envelope to capture the seed of the Strong Variation
✅- Start a new prompt tree with the final image of the old tree, adding the seed of the strong variation
✅- Zoom out without remixing
✅- Vary the Different regions of the output depending on intention, in this case I prompted "a reflective surface of the lakeside" for the ground and a "lava forest" for the background

In total, this uses a combination of scaffolding, disruptions, seed reinforcement, zooming, and regional variations to create the final result.

Your browser does not support the video tag.

 Here's an example of seed reinforcement in action. The basic idea is that the initial combination of the images (i1 + i2) used in the prompt and the prompt itself are going to create the initial grid (i3) onto the seed of its choice (i3), meaning that Midjourney has effectively consolidated all of this information. Put into a sort of equation, it would look like this: s[(i1+i2)p] = i3 Assuming that Image 1 is the sketch...
 
 
And image 2 is the vector...
 
 
...and using the prompt large tyrannosaurus rex, devil dinosaur,
venomized unbiased sketch, visible scaffold lines,
centered, by Frank Frazetta it'll give this result
 
 
This is great, because now all I have to do is replace the image of the sketch with a result that the seed (3732560190) itself produced, which will make more coherent and perhaps better result than what Midjourney could produce on top of the initial sketch.
 
 
- Changing image quality from inpainting entire images
 
use THE WHOLE AREA for inpainting.
 
Your browser does not support the video tag.
 
 
 - Fixing words with inpainting
 
 Midjourney had a handicap: text generation. The words it generated were mostly meaningless.
The new inpainting feature solves this problem.
This guide will show you how to do it:
 
 

In this tutorial, I'll be showing you how to:

1. Use Midjourney inpainting to make meaningful words
2. Take advantage of word spacing for best results
3. Change the font of the text
4. Avoid a common problem

This technique really opens up so many possibilities.

Let's dive in!

Example:

Here's a sign generated by Midjourney with the following prompt:

C O F F E E sign in front of a cozy coffee shop, clean, minimalistic, foliage, natural light

The coffee sign is clearly misspelled.

Let's fix it.

Using inpainting, select the region of the sign with spelling mistakes.

Then, replace the prompt in the text box with the appropriate letters and press submit.

Here are the results.

Not bad, but we are still missing an 'E'.

We will continue carrying out inpainting to add the extra 'E'.

The question is, which of the 4 signs below should we perform inpainting on?

It turns out you should only perform inpainting on the 4th sign.

We shouldn't do inpainting on signs 1, 2 & 3.

This is because there isn't enough space for Midjourney to insert the letter 'E' at the end while keeping both sides balanced.

Even if we try, no 'E' will be added.

After performing inpainting one more time, we have our COFFEE sign spelled correctly!

Now, here are 2 more useful tips:

I'll show you:

- How to change the font of a word
- How to avoid a common problem

We will be using the following example, where I want to change the word "FAMARE" generated by Midjourney into "FLAME".

 

- Inpainting with any image input

Picture in a picture? Let's take a look at using an image prompt with Midjourney's new inpainting feature!

First, I need a starting image. For this example, I have chosen a woman wearing a plain white t-shirt. Next, I need my reference image's URL. (See MJ's guide to image prompts for help with this.)

From my starting image, I use "🖌️Vary (Region)" and select a portion of the t-shirt. Then (with Remix Mode on) I add the URL first, and then a word prompt. In this case, I used the original prompt that produced the image of the flower illustration.
(If you are using a non-MJ image, you will need to write a new word prompt.)

Ta-da! Now her shirt has a pretty graphic on it. Have you used image prompts with inpainting yet? How were your results?


a young woman with sunglasses and dark hair, wearing a plain white t-shirt and dark denim jeans --style raw --ar 4:3


flowers & art project, in the style of letterboxing, childs drawing, hand-drawn animation, light brown and purple, simple, colorful illustrations, primary colors, dark orange and green --ar 4:3


*selecting the area on the t-shirt and adding the prompt*



https://s.mj.run/-kW0Mh93dJY a white t-shirt with a graphic showing flowers & art project, in the style of letterboxing, childs drawing, hand-drawn animation, light brown and purple, simple, colorful illustrations, primary colors, dark orange and green --ar 4:3

   - Creating large scale changes to a single generation

 

MidJourney Inpainting Demonstration - Part 2

This I time I change everything in the image.

Watch as Harry Potter works his magic and transforms in to Goku.

Follow for more, and consider a like/share if you enjoyed.

Your browser does not support the video tag.

 

 Found this article useful? Share it with your friends

For donation, contact or follow https://khalil-shreateh.com/links/

 
Previous article: How to use the best latest iOS 17 features Prev Next article: How to Verify Your TikTok Account and Benefits of Getting the Blue Checkmark Next
  • How to Add a Website to Your AdSense Account Linked to YouTube
  • How to Turn Your Photos into Studio Ghibli with ChatGPT
  • Transform Any Image into The Simpsons Style Using ChatGPT
  • Error 503 When Using Gemini The model is overloaded. Please try again later.
  • Remove copyright from video “Erase Song” AI Feature

 

Eye-catching Slogan!