Midjourney – the SECRET to training Midjourney AI

By | November 19, 2022

Is it possible to train Midjourney? so you can create the same character from different angles and perspectives. This is how to do it. The secret to training Midjourney AI, stable diffusion tutorial. And how to do character design with the help of Midjourney, DALL-E and Photoshop.

Can you train my journey I mean train it So that it can create the same character In different poses and from different Angles and if that’s really possible why Aren’t everybody else doing it [Music] Many of you have commented requested and Asked questions so thank you very much For that your wish list is long and a Lot of you wonderful people have asked About the possibility of somehow Training mid-journey in what results you Want or training it in different Characters or styles that you want to Create and we’ll dive deeper into that In today’s episode also we’ll be looking At how to get the right crop of your Image how to get a full body view for a Character for example or the right angle And so on we will also storm right Through 3D space shortly because there Have been questions about an easy way to Turn some of your mid-journey designs Into 3D models remember there are links In the description below among other Things to a free pdf cheat sheet where I Share some of my prompts with you so Check that out so getting started with My journey ninji journey or stable Diffusion is going into a rabbit hole I Have to admit that I sometimes feel a Bit like Alice in Wonderland And the more I learn the more humble I Feel because I may be getting wiser all

The time but I am also confused the Development of AI is so fast at the Moment what AI is the best for my needs What AI can I get the best results with And most importantly how can I train AI To create the characters and styles that I want to create so in stable diffusion For instance you can sort of train the Engine by feeding it without 20 Different photos of the same person in Different crops and in in different Perspectives and thereby you enable Stable diffusion to collect information And create images drawn from that Information so it’s light training but It’s not really training and I’ll Dedicate an entire upcoming episode on Stable diffusion so hit that subscribe And Bell button to not miss that episode But is that possible to achieve with Mid-journey yes and no to understand What’s going on we need to look at how My journey creates the images Ask the guidelines for mid-journey State You cannot feed images directly into my Journey like that due to concerns about Community public content instead my Journey lets you use images as Inspiration the image to image feature Usually along with text to guide the Generation of an image by the way image Prompting is not the same as building on Top off or initializing an image so the Way you work with the development of a

Character in my journey is a bit Different and as I explained in this Video about character design I am not All that Keen on the image to image Feature because it doesn’t really get me The results I want I get way better Results with comic book characters by Using a term like lookalike and then Using seed same seed and remix settings First let’s see if we can get my journey To give us the cropping and content we Want the right part of the image so to Speak so here I have to get 8-bit artsy Nerdy with you because midjourney it Needs to understand what it is you want And to be able to do that you have to be Rather specific about what you write in Your prompts now it’s not a guarantee That you’ll get what you want but there Is a much better chance of you getting a Decent result if you’re prompting the Right words Because artificial intelligence it can’t Read your mind you have to use words so Let’s just take a quick look at what you Call the different shots you need to be Asking for in a prompt I am going to Give you seven different shots to choose From there are a lot more and you can Just Google that but for now we’ll take A look at seven now in the description Below I mentioned a cheat sheet and that Has also got these different camera Shots and angles explained so check the

Link below to download the free pdf Cheat sheet now first we have the full Shot the medium full shot Cowboy shot Medium shot medium close-up close-up and Extreme close-up and then you also have To choose your angle so it’s like if It’s a high angle or a lower angle like First view or frock view also when it Comes to getting the right results with My journey you might want to look at Different perspectives and we’ll take a Look at that in an upcoming episode so We will also look at style expressions And what styles to prompt for so hit That subscribe and Bell button if that Sounds interesting to you So now you know the basics of what you Need to ask me Journey for but sometimes Mid-journey will still crop your Character in a weird way the trick here Is to actually mention in what angle you See the face in for instance and I’ve Had good results with mentioning details On the eyes too so that mid-journey Doesn’t mess up the eyes because even Though mid Journey 4 is way better at Eyes than version 3 it still tends to Mess up both hands and eyes a little so If you’re wondering how to fix that well You should check out this video that I Did on that subject but back to Mid Journey and how to get the face in the Image and not to get it half chopped off Because mid-journey can still struggle a

Bit with good compositions even though It actually does a really decent job at Creating good compositions so the trick Is here as I said to mention details on The face or the hair or the angle of the Head like this so Here we’re just gonna jump into the seed And remix Expressions too because we Need those in order to get what we want To get the remix settings toggled on you Simply write slash settings and then you Toggle on remix Now let me show you what I mean so let’s Say I want to create a comic book Character a viking with long blonde hair And metal armor so that’s what I’m Prompting but the results are not Exactly what I wanted There is something about the image that I like though for instance the colors Are great and the feel of the image and The character is really good but he’s Cropped all wrong and I can’t see his Face So I’ll get the seed off that prompt and Then I’ll simply do another version Where I change the prompt but use the Seed The term seed sets the seed for an image Symmetronic creates images from Diffusion or noise so when you use the Term seed it means that mid-journey will Use the same noise or diffusion to Create your image from the same sort of

Seed And using seed can sometimes help keep Things more steady and easier to Replicate when trying to generate a Similar prompt again or when you’re Trying to get the same sort of image but It can also be used for creating the Same character in different poses from Different angles and perspectives Yes You heard me right Pretty darn awesome so why aren’t Everybody else doing this well they Probably will be doing it after they’ve Seen this video but let’s face it Artificial intelligence is not an easy Tool to use and many people simply don’t Have the time or energy to sit down and Learn how to use it properly so many People think that they can just write a Prompt and then when it doesn’t turn out The way they want the abandoned ship so Let’s dive deeper into the seed Expression now how do you find the seed For the image you want then well you add A reaction to that image that you want To get the seed from and you do this by Right clicking on the image and then you Press add reaction and then other Reaction and here you just start by Writing and and that’s going to bring up The envelope now my journey will send You a private message and here you get The seed number so now you can go into Versions and tweak your prompt and add

The seed expression or you can create a New prompt with that seed so now I Prompt that my Viking should look like Chris Hemsworth as I explained in this Video about character design well you Tweak mid-journey by adding what Celebrity has the look you’re going for I change a bit in the description and Then I add the seed number As you can tell all of a sudden I have a Character to work with and it’s pretty Darn Magic Same scene so what is the same C well You use the term same seed to affect all Images in the resulting Grid in the same Way if you don’t use same seed each Image in the grid will use a different Slice of the Original Seed meaning that Your four versions will look differently And more varied but if you use same seed The four Images will use the same slice And in this way as you can see I Actually get a Viking that sort of looks The same and I can now use Daly’s paint Out tool and Photoshop to do corrections And yes if you want to learn more about Dally and the paint Out tool well I have Got an upcoming episode on that topic And as I said in the beginning of this Video the world of artificial Intelligence and art generation is Moving quickly and therefore I have also Got an upcoming episode where I will Look at nvidia’s nvidia’s new AI engine

In development and stable diffusion 2. And we’re we’re also going to take a Closer look at the impressive energy Journey for anime style comic book Characters and animation characters so Hit that subscribe and Bell button if You think that sounds interesting and You don’t want to miss out when I drop That episode So let’s just talk shortly about 3D Because is there an easier way to turn Your designs into 3D models the fast Answer yes and no uh well Of course Adobe has some amazing tools To create 3D models but they’re not easy To use and in general 3D has just like Any other cool thing you want to learn a Steep learning curve but of course there Are always a shortcut or two that you Can take so if you go into Monstermash.zone see the link in the Description below and upload a 2D image You can actually turn that into a 3D Model in a fairly simple way so I’ll Just grab an image of a crocodile from The side and upload that then I’m going To draw on my image first the body shape And then the legs and then I inflate the Shape now to get all four legs off your Crocodile well I need to double click The legs I’ve created to copy them and Then I click animate and now I can Download a glb file that I can import Into blender omaya whatever application

I’m using for 3D and blender is still Free I think as far as I know let me Note that working in 3D mesh is not easy You need to get to know the 3D workspace And 3D apps are complicated but now you Can import your file and play with it Let me also note that one of the hardest Things to create a 3D model from is a Face because you can’t just inflate that You need to put in the eyes and the Sockets and what about teeth and the Tongue and oh my God it’s so complicated But if if you do want to start playing With coloring 3D objects well procreate Actually has some cool 3D models that You can download to your tablet and you Can start drawing enough about 3D let’s Get back to Mid Journey because I just Want to show you something called clip Interrogator so check the link in the Description below for that So what you do here is that you upload An image that kinda looks like the one You want to create with my journey or a Stable diffusion and then you drop the Image in here and it’s going to come up With a good prompt for that so if you’re Starting from scratch and you want to Learn more about prompting well you Should try this out and also I’ve got a Video on that topic that you might find Interesting then I just want to mention We have hit the first thousand Subscribers thank you very much for that

I really appreciate it and I’ve been Doing my own little happy dance here at This video remember creativity doesn’t Wait for the perfect moment Carpe Diem Stop procrastinating and seize the day