Transcripts
1. MidJourney Boundless Creativity!: Welcome to the Mid Journey
AI Image Generation course. Your gateway to
unlocking the limitless, creative possibilities of
artificial intelligence. And this course will take you on an exciting journey through
the world of Mid Journey, a cutting edge software
that empowers you to create stunning
images using AI. Our course begins
with the basics, ensuring that you're
comfortable navigating the mid journey user
interface right on discord. Whether you're a beginner
or an experienced user, you'll find value in mastering the core functionalities
of this incredible tool. We'll dive into the art of
crafting effective fronts, teaching new techniques
that will make your AI generated
images truly unique. Learn how to coax mid journey into bringing your
ideas to life. Just the right words, but
that's just the beginning. We show you how to remix prompts and even add custom weights, giving you full control
over the creative process. I'll unleash your
imagination and let mid journey adapt to
your artistic vision. Ready to take your skills
to the next level. Our course doesn't
stop at the basics. You'll cover advanced
prompting tricks that will set you apart. As a mid journey master, be prepared to be
amazed by what you can achieve to help you put your
new found knowledge to use. We've designed four
captivating projects, including web design, consistent character generation, and other real
world applications. Learn how to
seamlessly integrate mid journey into your
daily work and activities. Making it an invaluable tool
in your creative arsenal. So whether you're an
aspiring artist, a designer, or someone with a passion
for AI driven creativity, this course is your key to unlocking the full
potential of mid journey. Join us on this
exciting journey and embark on a mid
journey of your own. Enroll now and get ready
to create, innovate, and inspire with the magic
of artificial intelligence. Mid journey, AI,
image generation tors where creativity
knows no bounds.
2. Setup discord: Start using Mid Journey, we need to set up
our Discord account. Discord is an application that allows you to create
and engage with communities online and several
services like Discord. Instead of creating
their own platforms, they use Discord to allow their users easy access
to their services, which in our case is
the mid journey AI bot. We'll use Discord to
create our images. First, let's go to Google and search Discord and click
on the first link, and then click on Login
and then register. Now we're at the account
creation page here. You just need to give your
details, E mail address, username, etcetera, And
then click on Continue. Now we have to complete
the silly checks and verify ourselves, and our account will be
created in just a few seconds. Now our Discord account
has been created. Just a couple more things to do. We need to go on Google
Search Discord again, and click on the first link. And either download it or use it directly
from the browser. The third step is to
search for mid journey. Click on the first link
to go to the home page, and then click Sign In and
then authorize right here. Now we've successfully
signed into mid journey. The fourth thing that we need
to do is purchase a plan. And we need to purchase
a plan as there is no free version of mid journey. The basic plan should suffice. And you purchase it
like you would purchase thing on line after
the purchase is done, we can start with the
image generation. That's it. Those are the
four things you need to do before we begin,
just to reiterate. Number one, create
a discord account. Two, open discord in the
browser or download it. Three, sign into the journey. And four, purchase a plan.
3. Create First Image: We have our Discord open here, and the first thing that we
will do is we will go to our mid journey channel and select any chat
room that we want. I'll come down here and type, imagine, press the Space bar. Now, whatever type of AI
image you want to generate, you just have to describe
that image in words. Let's say I want to generate
a cat wearing a hat. I'll type cat varying hat. Then I'll press Enter
because this is our first time images
using mid journey. We'll have to accept
the Terms of Service. Click on Accept US, and here you have mid journey with these
four generations. It took about 15 seconds
and I cut that part out. As you can see, we have
these four variations. The top left one is image on the top right is image to the
bottom left is image three. The bottom right is image four and I'll tell
you why that is relevant. At the bottom, you can
see an array of buttons. You'll see 1,234,123.4 We also have a ninth
button which will pick, which will make it just a bit. What is the meaning of
and what's the meaning of means upscale and
means variation. For example, let's say I
like image number one, which is on the top left,
That particular one, and I want to
upscale that image, I'll simply click on One Mid journey will
generate the same image, but with a higher resolution. It's more suitable
for online use. This is how we can select
an image and upscale it. Now let's create different
variations of our, let's say I want to create
variations for image four, the bottom right one. I click on four. After clicking on four, you'll see these four
different images in the same style of that fourth image that
we wanted variations of. We use this button to guide mid journey in the
artistic style we want our images to have.
4. Remix and Seed: Now let's discuss re mix mode that is used
to change prompts, parameters, and aspect ratios between different
iterations of images. It will create a new image that resembles your initial image. Which will help you take away or add anything that
you initially forgot. On top of that, you can change
the setting, the lighting. You can transform the
subject or create a more complicated scene using a previously generated
images image. It's similar to the parameter which will get into
in just a bit. To use remix mode, we can just turn it on in two
different ways. Type prefer remix, press
Enter, and now turn On. Another way to turn it on
is by typing Settings, press and here at the bottom you can
toggle on Re Mixed Mode. I'm going to go
ahead and dismiss this message now let's use it. I'm going to imagine an
image, Gothic style, mid century home in the
middle of the desert, and press enter the
image that it output. This is the one in to
initiate remix mode, we're going to click here at the bottom on image
number three. Click on three, and
now you can see a new prompt has come
up, the remix mode. Let's say I want to add
people walking around. Take that in and click Submit. An image that is very similar to the one that
I chose in the beginning. But now my Journey has remixed it so that there's more
people walking around. You can also re mix
an upscaled image. Let's choose number three. Again, three upscale it. Now if you click
Make Variations, you can now make variations
to an upscale image. Now let's discuss the parameter, which is a random
number associated with every image I generates. Every image is given a
number by mid journey. The parameter initially
sounds pretty confusing, but it's extremely useful and
can be used in many ways. Every prompt you run will be
given a number by default. You can choose a number
yourself by using the parameter in your initial prompt
to use it Seed, and then press Enter and
type in a Seed number. Also, I'm going to
use 123 for this. I have the prompt
followed by Seed and space 123 and press Enter. If you run a job with the same prompt and
the same seed number, then it will come out
exactly the same. If you change the text
in the prompt slightly, we'll try to adjust the image for you according
to what you change. This is useful when you create an image of something
you really like, but you don't like
certain aspects of it, like the color or the weather, or maybe you just want to
add some more people to it. It's very similar to
the variation command, but with this you
have control over what is being adjusted
rather than journey. Making the adjustments for you is slightly different
than re mix because a remix can only be applied to an image that you
have already made and it will create a new one
off of your adjustment. A adjustment can be
applied to any image that you have made in the
past, present, and future. I mean, you can use the
same set anytime you want to generate an image
to get a particular style.
5. Panning an Image: All right, in this one
I'm going to show you the pan tool which comes
with version five. You can pan up down
left and right, meaning you can add to the
image from all four sides. Let me show you what
I'm talking about here. This is an image of a
bulldog reading a book. And then instead of creating variations to add to the image, you'll see these four arrows. You can go ahead and pan in
this I selected pan, right? And you'll see four variations of what could be
added to the right. Like moving the
camera to the right. It interprets what could
be on the right side. The dog is pushed to
the left of the frame. And then there's now
another book with a lamp. This next one you'll
see is panned up. Now there's a light
and a book or two levitating in the air. There you go. This is how
you go ahead and do panning. I'll go forward slash imagine, and I'll show you how to do it. I'll just go do reading Enter. And when it's done, I'll show
you the buttons in action. All right, here we go. This is what a dog reading
a book looks like. And it gave us four examples. This bottom left one
is of a bulldog. So let's go with
that. What do you do? I would click on Three to upscale it once the
upscaling is done. And now we've got very
strong, very subtle. We have different zooms,
we can square it. Now we've got four
arrows to pan it. In this case, I'm going
to pan it to the right. You can go ahead and
use remix as well. Do reading, if you want to add something in, you
can go ahead and do that. I'm going to submit
it literally, that's all there is to it.
6. Prompt Weights: So now we're going to look at
mid journey prompt weight. This is another tool for your prompting skills
that you need. And as you may already know, mid journey is all
about prompting and writing a good prompt book, give you the image
you're looking for, that you're looking for.
So let's get into it. What are prompt weights? Well, if you're looking
for one part of your prompt to be more influential on the
mid journey bought, then you need to
add prompt weights. If you want more of, you
can use prompt weights. Even if you want less something, you can still use
negative prompt weights. So we're just going to start
with a simple prompt here. I'm going to go forward
slash imagine space, and then let's go with
tall building blocks. So tall building blocks. And now I'm going to
give it an aspect ratio of let's say nine by 16. Okay? Nine column 16. Now if you don't know what aspect ratios are,
you don't need a wiry. We'll do a separate
video on that. This one is all about
prompt weights. This example right here is without prompt weights
and it's still loading. It's getting somewhere, okay? Okay, let's take a
look at our results. Okay, a lot more
toys feel to it. It's confused between
do I want blocks, toy blocks stacked together. Or do I want a building that's
made up of similar blocks? It's mainly making a building
that's made of blocks. In some of these
images on the top, that definitely looks
more like toy blocks. On the top right, it's like
a toy block and a building. Things that we might find in
buildings are like windows, roofs, and we're not seeing
that here on the lower left. It's stacking some
things together. A lot less architectural
feel is being provided here. We can help mid journey out
with some prompt weights. Let's look further into
that for the next prop. Let's go forward slash
imagine and space. I've been told building
and, okay, tall building. To specify the prompt weight, we go double column and
then prompt weight. So let's say I want to double the emphasis on the
tall building part, so I'm going to just
go with two, right? Because I want to double
it and now we have blocks, double column space with
double column and one. It's not necessary
to specify the one because the weight
is one by default. But I want to just show it to you visually that we're doubling the emphasis on
the tall building. We're keeping it default
on the blocks part. Now let's go ahead and put in our double dash R nine by 16. As you can see here, we have more emphasis on
the toll building than we have on the blocks. Definitely, that
is the case here, because now it has a more
architectural feel to it. The building is
going up straight. There's not like a mess of blocks at the bottom and
then some blocks at the top. We definitely have
a building here and there's definitely
more windows involved. And image three looks real compared to the other
three. That's really good. I also like the first image
looks more often anime one, but these are definitely
building blocks. All right, we can go even
further now with the weight, but you need to note that even giving it a
weight of 100 is going to be similar to giving it a weight of, let's say 5567. That's maybe the
maximum emphasis. You can give a
word. I'm going to pace in the previous
prompt and give it five on the tall building and keep it the same
with the blocks. Let's see what that looks like. Okay, definitely
loading. All right. All right, I like them left one. It kind of looks
like a picture taken from the bottom of a skyscraper, while the others are
a bit different. You know, they're taking at the mid level of the building. But you know, mid journey
is still fighting with, you know, between
buildings and blocks. So you might want
to try that out. Try a couple of prompts out. Try a couple of different
weights and we're also getting a lot of clouds
in our prompts here. And some trees in the
case of the top left one. You know what? Let's say we don't want trees, we're going to have to use
negative prompt weights. Now, negative prompt weights are different from
negative prompts. We're going to cover negative prompt at a different radio. Now we're just going to
give a prompt weight to, let's say the word
clouds and trees forward slash imagine space. And then paste it past the original prompt with the five to one ratio to tall
buildings and blocks. I'm going to go to
trees with a weight of negative, negative one, okay? Okay, then we have clouds with a weight of let's
say negative three. All right, Okay, keep the aspect ratio
the same because we want a tall picture, it's
definitely loading. It's going to give us something. Let's see what that
is. All right, now this is a lot better. This is a tall building made up of blocks with
a clear background. No clouds, definitely no clouds. We can see traces of clouds
in the bottom two images. Because we gave it a
negative prompt weight. It's lessening the
significance of clouds. It's definitely trying to. We have a lot better
images, let's say. If we didn't want clouds,
these are better. In that case, we have four buildings and they're
all made up of blocks. That's all there is to
majority prompt weights.
7. Aspect Ratios: I wanted to make a
quick video about aspect ratios in this
version five of mid journey. This version five of mid journey can handle
any resolution. We no longer have
the limitations that we had with four if you've been using it for a while now and then the
versions before it. Let's first let's
look into how we can use this aspect
ratio version. We go imagine type in the
prompt like we would with any other version and
then we go AR 16 by nine. You can also go tall with
two by three or even super wide with let's say the aspect ratio
of four by one. This is the panorama ratio and backs really wide and you couldn't do that in
the previous versions. And let's see what we
can get up to with this. The Mid Journey has become
a lot more ratio ware, which means it can
adapt the designs to the ratios you give in it
much more effectively. So let's take a look at an
illustration in nine by six and then 16 by nine
ratios to see how this works. You'll notice how the
figures are drawn differently, these
ratios perfectly. So let me tell you an issue that was there for four
weeks and then 16 by nine ratios to see how this works was a
cropping problem. So when you specified aspect ratios that were different than the ones
that the A I wanted, usually find the
photos to be cropped down to fit the aspect
ratio you asked for. And you can still get that with this version, but
it's a lot better. Let's talk about how this version has
improved some things. While version four
couldn't give us the exact 16 by nine ratio, we needed it switched to the
seven by four ratio instead. That meant you
couldn't just send the image straight to the frame. You couldn't fit it because
it was seven by four. It was closed but not exact. You'd have to drop
it, But version five gets it really accurate. Really accurate. You still have to tweak it by just a bit. So 16 by nine might sometimes mean 16 by 8.5 or
something than that, but it's really close. Okay. There's also some scaling, but we'll touch on that later. For now, let's go through some common aspect ratios and their best uses
and some examples. Four by five, which
is also eight by ten, is great for social media, particularly Instagram
and Twitter, because they fill up a lot
of the screen on phones. It's also used for picture
sizes quite a lot, and prints also come
in those sizes. So it's really easy to buy frames for this
four by five size. Two by three is another
common photo and print size. It's often used for posters. Again, very easy to
get frames for it. Nine by six is store and
perfect for phone screens, while 16 by nine is the go to ratio for TV's and wide
screen digital displays. And that is it for this video.
8. Negative Prompts: Okay, there's a
difference between negative prompts and
negative prompt weights. Negative weights lessen
the significance what a negative prompt removes
that thing completely. Let's go forward.
Imagine the space and then type in what you want. For example, I just do
a male face then no. For the negative prompt, I don't want hair,
eyebrows, hands. Make sure you put
****** in between those because the commas
really don't matter. You will need to put in ****** between prompt,
prompt features now. Okay? Okay, so we're definitely
getting no eyebrows, no hands, and no hair. Right. Arguably, there is some eyebrows involved
in these images, but that could be because of, you know, the
training date itself. You just try out face with no eyebrows for
a different prompt, and I see if that works better. But for now, this
is how negative prompts work with a no command.
9. Advanced Prompts: All right, so for this lecture we're going
to go through blending, multi prompting, and
some other things, advanced tricks in my journey. These are some amazing commands and techniques that
we're going to go over and they'll make you a
better prompting engineer. We'll start out by generating
two different images. For the first image, I
wrote Samurai under Saka, and I got these results. All right, now for
the second image, I wrote Hulk contemplating, and I got these
results. All right. Now, interested in
creating a Hulk from a parallel universe
where he is a samurai, there is two different methods. The first one is by blending the images using the
blend command or simply uploading the URL of the
first image followed by a dash and the second
image in the first prompt. This method is pretty
straightforward and you just need to
make these two images like so we call these results
using this method. Okay? It doesn't matter which
image you put in first. In the prompt, when
using the blend command, you could interchange them and have only a
minimal difference. But that's because if you haven't specified
the seed number, if you also specify
the seed number, they'll be exactly the same. Let's look at another
example generated here of a waterfall blended with
an image of a mountain. To get this result that is pretty simple and amazing
at the same time, let's move over to
the second method. We'll use a command
called Describe. What most people don't know is that mid journey has a built in function that describes
any image you want. You just go described and describe and attach the link
to the image that you want. Mid journey to describe. Let's use the describe
command on our two images. And let's see what we get. Describe, upload
the image and here we have the link and put
the command to test. We get these four
descriptions of the image based on how mid
journey understands the image. And it does a pretty
good job to be fair. We can use one of these
descriptions of the image and generate one and compare
it to the initial one. We can simply click
which one we want to choose to generate our image. So we'll choose the fourth
description, which is here, The Incredible Hulk
by Pauline Hansen and blah, blah, blah, blah. Anyways, okay, so let's
go click on four. It gives us the
option to change. The prompt clicks a bit, and these are pretty
good results. It turned out to be amazing.
I think it nailed it. It got the same vive as the original image as we can see here on the left
to right comparison. Now we can do the same
with the samurai image. We can write describe. We
upload the image and hit Enter. We can choose one of the
descriptions that it gives us and generate four
new images like so. And we get another
incredible set of images. And, you know, maturing is
really good at describing images and reverse
prompting as some call it. So instead of using the blend
command mixed two images, now we'll use an advanced
prompting technique called multiprompting's simple, but it requires a
lot more effort. Now use the described command
for both of these images, and you select the prompt
that you like best. And now we'll combine those
two prompts together. So we'll put in the
first prompt, the Hulk. And then we'll put in the
second prompt of the Samurai. Let's head over to mid journey, separate the two descriptions by this double holding here. And all we have to do is
wait for mid journey. These results, I have
to say, are amazing. I generate a few
different sets of images. This is what we get here. And I'll say this is giving us a lot better results
from the blend feature. That can also vary
from case to case. But in this particular case, the describe and then
multiprompt work better than the blend feature. This is where we come to the prompt weights and the different aspect
ratios of the image. This is just a way to tell our AI what aspects of
the image to focus on. This helps you take your
prompt engineering skills to a different level souls
straight from the Mjourneydcs. There is something called
multi prompt basics. It says that adding
a double column to a prompt indicates to the journey bot that
it should consider each part of the
prompt separately. In this example below, for the prompt hot
doc, as you see here, it's considered as one word and journey bot produces
images of tasty hot dogs. If the prompt is separated into two parts hot then double
colon as we've talked about. Then do both concepts are
considered separately, creating a picture of
a dog that is warm. Moving on to Prompt weight, when a double colon is used to separate a prompt
into different parts, and we've covered
this previously. You can add a number
immediately after the double colon to assign the relative importance to
that part of the prompt. Here's an example, another one, Prompt, hot, double colon dog. As we can see here, it produces a dog that is hot
or that is warm. And if we change the
prompt to double colon and adding a number two after the word as we
did in the previous, well then dog makes the word
hot twice as important. It produces an image of a
dog that is even more hot, Hot, double colon, Do hot
double colon number two, do. These are two separate pumps
and they make the word hot, The prompt twice as important.
Here's another way. Prompts can be used if
we use negative weights, which we've also
covered recently. In this example, this prompt, vibrant tu fields has been used. As a result, a range of color
tulips has been produced. Let's say you want to remove
the red colored flowers or have less red color. In the image I can write. Imagine, prompt, then
vibrant tulip fields. Double column red,
double column five. Here are the results with
the negative prompt weights. By doing that, I can
control some parts of the output and get
these amazing results. Here's a comparison
with the first image. Without the negative prompt, we can see that we've
got similar results, but without the red
flowers or the color, or the color red in the image. That will be it for this video.
10. Some Tips: We're just going to
go over a couple of tips and tricks that you need
to know when using journey, and it just takes your
prompting to the next level. The first one of those tips is adjusting the chaos parameter. Now when adjusting
the chaos parameter, you can select the value 0-100 And what the chaos
parameter does is it determines how varied
your initial four images are that are appearing
with the grid. If you choose a very low chaos, it will make all four of those images image
is very similar. If you have a chaos of 100, then all four will vary greatly. In order to type out
the chaos parameter, what we want to do is
space and then the value. So I'm going to go
for 50 halfway up the scale just to see
how much they differ. So as you can see,
using the chaos of 50 has changed these
images dramatically. Each image is nowhere
near the same. Another tip is to try single word prompts for
surprising outcomes. Just like any other
prompt, we can start out by typing
imaginatingabnjustypele word. One of my favorite single
word prompts is surrealism. And you can still add parameters here if you want the quality of, let's say, a chaos of 50. You can do that and change
the quality as well. But just adding one word lets my journey run around
with its imagination. As you can see,
we've come up with a lot of very
interesting images. You can look at these too,
they're very different. Another tip is to specify the perspective you
want for your image. Right now I have this prompt for a photo realistic astronaut
walking on planet, red dust. And then I specify what
perspective I want. If I want an aerial view
that I'm going to do, this is the result it
gives for the aerial view. It will be above the subject or the object that you want
to have with your prompt. But when I change the view from aerial view to
worm's eye view, what you'll notice is it has a beneath shot of the subject. That's why it's very important
to specify what kind of perspective you want within your images to get the exact
picture you're looking for. Another tip is to specify any lighting
conditions that you want. What kind of lighting do
you want your image to be? Is it golden, hour, twilight side lighting, silhouettes,
morning light? You've got to be specific if you want the light coming in from a certain angle in order
to get that perfect image. That's just another one of those components that you
have to add to the mix. We take this prompt,
for example, and add golden R
to the end of it. What you'll notice is it changes the image
to look like this. Instead of looking like this, without the specifying
lighting conditions. This is without
specifying lighting. And this is with
specifying lighting, it has a whole element
to all of your pictures. And another dip is to specify the desired
composition of your picture. One of my favorite
compositions to use is the Rule of
Thirds composition. So as you can see right now, this cowboy is centered in
the middle of the picture. And this is because
I did not specify what composition I
wanted for this photo. So if you notice, I'm
going to add in rule of thirds and watch where
it places him in the photo. Now as you can see, instead of placing the subject
directly in the center, it moves him to the
right or the left, giving the photo
room to breathe. And instead of looking
at the camera now, he's now looking off towards
the open space in the image, in this bottom left one. This kind of gives a dramatic effect because there could be something maybe coming behind him and he's looking
off this way. Adding composition to images or videos changes
the entire wipe. And another tip to mention is the desired
camera settings. Now this especially works well when you're trying to generate a hyper
realistic image. Where you want it to look
like this picture was taken. In real life, things you should be specifying are focal length, your aperture, your
camera, your ****. And this will really
help journey decide what kind of image it's generating. Let me
show you an example. So as you can see in
this recent prompt, what I did was I use
a Nikon 850 with an 85 milliliter ****
and an F stop of 1.8 and that made the
images turn out very cool. It has a very shallow
depth of field. It's high quality, It looks like it was taken
with an actual camera. But what if I wanted this to
be taken on a film camera? Well, then I would
have to mention that. So now what I've done is
I've changed this prompt. In order to cooperate
the film camera, I've added the Kodak
ultra nine with some film grain and check out the differences in settings. As you can see, the
results are now like this. Just by changing the camera
and a few other settings, it made it look a lot older. Just by changing a couple
of camera settings, it's completely changed
the outcome of the image. That will be it for this video.
11. Midjourney Copyrights: On the Mid Journey Gid
Book instruction manual, there is a section on copyright and trademark rights you
give to Mid Journey. And they say by
using the services you grant to Mid
Journey, its successors, and assigns a
perpetual worldwide, non exclusive sub
licensable, no charge, royalty free, irrevocable
copyright license to reproduce, prepare derivative works
of publicly display, publicly perform sub license and distribute text and
image prompts you input into the services or assets produced by the service
at your direction. This license survives
termination of the agreement by any party for any reason. So this, in my understanding, means that Mid Journey is allowed to do whatever
it likes with your work, that it is allowed to
prepare derivative works, publicly displayed,
and do whatever it likes with it, essentially. However, your rights are
outlined in the next section, and this is a little bit
more comforting Subject to the above license, you own all assets
you create with these services provided they were created in accordance
with this agreement. This excludes upscaling
the images of others, which images remain owned by
the original asset creators. Journey makes no
representations or warranties with respect to the current
law that might apply to you. Please note, Mid Journey is an open community which
allows others to use and remix your images in prompts whenever they are posted
in a public setting. By default, your
images are publicly viewable and re mixable
as described above. You brought Mid Journey
a license to allow this. If you purchase a private plan, you may bypass some of these
public sharing defaults. So the key thing
here is that you own all the assets you
create with the services, meaning that you can
sell them commercially. So this does not apply
to non paid members. And also if you're
a corporate user, but you are using the plan
designated for single users, you're also forfeit from
much of the rights. So now one other thing to bear in mind when creating works in mid journey is that
you may stray too close to existing
trademarks and copyrights. So it's important to
note that even if you created an original work, if it does somehow
resemble something that exists in somebody
else's copyright, then you will be at fault. Another thing to bear in mind is you need to make sure
that you are allowed to use the images that you're using as a prompt inside
of mid journey. So if you're taking images
randomly from the Internet, be very careful because
it may come back to you. But I can recommend using
free resources like Unsplash.com forgetting copyright free images
as your prompt.
12. Tile Feature: Textures and patterns are an essential element
of any design project, and actually creating them
is a pretty difficult task. But with the mid journey
I image generator, you can easily
generate beautiful, high quality textures and patterns in just a
few simple steps. In this editorial,
I'm going to show you how to use mid
journey I image generator to create
seamless textures and patterns for your designs. With this knowledge, you
can create a vast array of digital artwork for
any craft application, head over to the
mid journey server, or go into the mid
journey direct messages. By default, mid journey is
set to the latest version. But if you want to
be able to create seamless textures and patterns, you need to make sure that
you're on version five. At the bottom, type
settings, press Enter, change it from M J version
four to version five. You can dismiss the
message and then we're going to use the tile parameter. And that's only available
in version five. You can really create a pattern or a texture of
anything that you want. But for now, let's try to
do something like type in, imagine space grass space tile, space R for aspect ratio. And let's do one to one
so that it's a square. And presenter not
only did Mid Journey create four different images
for us to choose from, but each one of
these is completely seamless and tileable.
In my opinion. A lot of these are very cartoon and I want to try
to get away from that a little bit and create something that looks
more realistic. This time I'm going
to type imagine space realistic brass tile and AR and then
one to one again. This will create a
seamless tile texture And it will make the
aspect ratio one to one, which is a square As
you can see with these, it's a little bit more
realistic looking. These are really zoomed in, so if we take one of
these images, upscale it, and then we can bring
it into Photoshop, we can start to see how it's a seamless texture
for this image group. I like the bottom left one. I'm going to go over
to upscale three. After it's done upscaling, I'm going to click the image, right click and do save Image. Now if you want to
check whether your generated tile image
of a one by one ratio, that's what we
call a tile image. If it's a seamless texture that it can be tiled side
by side or top to bottom. You can just search on line for seamless textures or seamless texture checker
and you can open this website and copy and paste your image here
uploaded and then it'll show you whether your generation is
actually seamless. For this example,
this image that I'm going to select
locally from my computer, this image is actually
very seamless. And you can see this here
on this online tool, that it's flowing from left
to right and top to bottom. There are no breaks
in the design, and that is what seamless means. This is a really good tool for checking your generated images, the mid journey discord, to get some inspirations
for your tiled images. You can search
Google for anything. Let's say hardwood
flooring or grass, or let's say rough matt
textures for home designs. You can search any
of those keywords and you'll find a plethora of images to scroll,
the ones you like. You can just copy and paste
into Discord to provide the journey bot with some inspirations of what
you would like to see. With the prompt, that
really helps to narrow down the images that you want to create with this journey AI bot.
13. Web Design With Midjourney: Okay, let's now get mid journey to design
our websites for us. And this video will go over the props that you can use to create these
beautiful websites, and let's dive into
it. Let's try this. A colorful living page for a
company selling popsicles, light mint, light pink, and light color scheme with
cool abstract textures. Version five AR nine by 16,
and this is what we get. And honestly, all of these
are pretty good options. It gives us some
good variations, some creative visuals,
good layouts. And you can pick and choose from any of
these really overall, this first drive was not bad. One thing that might be
missing is a bit more detail. Let's make one slight
variation on the style. How about this light mint, light pink, light
yellow color scheme? Abstract textures
like zebra and cheat the bread with that
slight adjustment. This is the result. Argue, we got more
detail in this, but the popsicles
don't look that good. Now let's just go with this
option for I think it will give a good start to designing
our popsicle website. Now when you're creating your own design prompts
inside my journey, there are a few things
that you should include. The first is, after you've
written the basic prompt, you want to include the
colors and textures and specify the styles
that you want to give a more accurate feel to have it looking exactly
like you want. Next up, you specify
the aspect ratio, so it knows what screen
resolution you want to target. All right. Now, the
next design prompt will be a new landing
page for a Batman movie. Very serious, intense Batman
sitting over Garment City. And color scheme with
small amounts of dark blue version, 5.9 by 16. And this is, we guess,
really good results. I'd say for a first attempt
this output is really good because it does a
good job of putting ideas into our minds of what the graphics
might look like, the layout, what the first
section could look like. This web design definitely matches the intense
nature of Batman. This first class
was very helpful. Let's go with another one. Let's, let's choose
the first option, as it has a nice landing page of Batman looking directly
at the call to action. The rest of the
sectional layouts were pretty solid and
immediately gave ideas that you could take and implement when actually
designing a website. And this really is the point
of a tool like Mid Journey. It's not going to design
a final product for you, but what it is going to do is give you an infinite amount of ideas based on industry
style, color palette ideas. And you can just
toss those things into mid journey
and it spits out these ideas that you can get started with
designing a website. Now let's say this
Patman website was a bit too serious
for your use case. So let's see what happens if in this prompt,
instead of serious, you put very silly,
unprofessional, out of shape Pixar
incredible style. And this is the result,
and I really like this. I think the illustrations
are clever. So let's take the fourth
one that looks like mist, incredible in a Batman suit. And just like that, you can
take the exact website, change the style,
and it's going to give a completely
different result. And so if we're looking for
something more playful, more serious, darker, lighter, no matter what it
is that you want, Mij is going to be able to get those ideas up and
running pretty quickly. So let's get into
the final prompt, which is a cryptocurrency landing page with green accents, gold coins, slanted sections, and check this out. So I do believe that a lot
of these were intricate, but it's better to have
these outputs be too much and too many rather
than not enough, right? Because you can
choose what you like. When you've chosen them, you can just get some
additional variations going with the one or
two you do prefer. And then just pick and choose the different
design elements and sections from those designs. Now in this case, there were
a couple of designs such as this one which were arguably more
professional and unique. The output that it gave
was influenced by the prompt specifying
slanted sections. We have graphs mimicking
analytics data, and this big graphic here
catches the eye overall, a professional looking website. And that too, with just
a couple of attempts. So I think it's a really good
tool to get started with, you know, bumping
ideas and, you know, playing with them and getting a really good feeling look of what we want the final product to look like before getting into the actual coding and Figma
tools and all of that stuff. So this really saves
us a lot of time and gives us really good
designs in my opinion.
14. Creating Icons With Midjourney: Okay, I icons. Let's take a look
at these icons in the community section
of Mid Journey. Some of these icons
have different styles. Some look like app icons, while others look minimalist. And we can also have a look at the prompts used
to generate these. Then we can use
them as reference later to generate our own. Let's have a look
at this one here. This looks like a good
version of web icons. There's another one here which looks like great app icons. And all these look pretty good. So what I'm going to do
now is create a prompt on mi journey through Discord. I want to start with a
really simple prompt. I want an envelope icon, something that you have in most websites for
sending messages. Also a home icon
for the home page, and maybe a user profile where you have like a
silhouette of a head. But here's the prompt
for an envelope icon. Let's use stylized 750, which adds a more style
to the general prompt. The higher the stylized value, the more creative journey gets. Here we are. This is a very simple design to begin with. Let's
do another problem. We're going to be more specific, in this case to
control the color more because most
envelopes aren't orange. Let's go with
envelopes with icons. Now the only difference is that we've added the letter
S. But what this means is now mid journey will start generating multiple
icons for variation. And we can have a
look at which style looks better and work
our way from there. On the top left, we have four different
types of envelopes, blue, pink, orange,
and brownish. On the bottom right, we have
more options for icons, but this time it's
as if you're getting an icon pack rather
than a single icon. This is where specifically making sure that the
prompt has what you want will play a big difference in what kind of
responses you get. Now these icons themselves
don't look like an end look. They look more like a briefcase. And this is just because
they're overs styled. So let's see if we can
just improve all of these. So if you want, let's
say a white background, because when we have
all of these icons, you normally have to pull them out into a Photoshop and
remove the background. So let's make it
easy for ourselves and since Mid Journey doesn't
produce vector icons, it only produces JPGs. Let's just go with the
white background for now, what I'm going to ask is a more simple envelope
against a white background. And let's give that a shot. White envelope on a
white background, maybe even a white envelope
icon on a white background. Let's also add in UI and UX. As soon as I add in more
and more into the prompt, it gets more complex too. Let's not. Or complicated, especially when we begin. But later on we can add things like minimalist or for example, outlines and other things. Here it's generating
the envelope icons. You can already tell
these are better, but they do look like app icons because they've
got that border around them, like that one on the top left. Well, this could be useful if you are generating an app icon. Sometimes you just want an
icon for the website itself. These wouldn't be
ideal for that case, so okay to generate a
new prompt this time, let's do a white envelope
icon with a white background and no will app as part
of that no sequence. This means that I don't
want an app icon. So it'll stay away from imagining something
similar to app icons. So let's see what
this does, okay? Okay, so it looks like
they're generating correctly. We can already tell
and this is good. It's exactly what the
prompt was aimed at. Consider adding no two
statements if something is popping up in the image that you just don't
want to appear. Now from these, some
of these icons for the envelope don't particularly
look like envelopes. Some of them have extra
lines or extra paper folds. So let's go with the one
on the bottom right. Let's make variations
by selecting four. We've got remix enabled, but let's not change
this prompt now. And here are the four new icons. All of these seem to
have correct folding. You can really pick any of these to generate a large image, to upscale it, and use
it as part of a website. So let's go with upscaling
version three of these variations because that's the most accurate
one in my opinion. And let's see what
it looks like. This is what it looks like
when pasted onto a website, and I think this looks pretty
good in terms of image. My journey is also very helpful
for generating icons for websites or apps and a lot of other things
that need icons.
15. Designing Mugs With Midjourney: Print on demand products
like mugs and shirts, et cetera, are very easily
designed with mid journey. We can use prompts like this one here to create these designs. Now the words white
background in this prompt are really important because we're going to put this
on a white mark. And you can use any other
design with a white background. So you can do
watercolor flowers, let's do flowers here. But it could be a cactus, it could be other things
on here that you like. Sunflowers, birds would be good to mushrooms,
whatever you want. Let's just use flowers now, because flowers do
really well for mugs. And I've seen similar
designs on websites like Red Bubble. So here it is. The tile parameter here is important because it gives us a repeating pattern
on our products. You can copy this prompt
or you can use your own and we have
the results here. You have the original
prompt at the top, four variations, and
then at the bottom, we have an Upscale
of the second image. And now we can use
this upscale image as one of the designs
for the marks. The next prompt is a frame
of beautiful spring flowers, so you can see it as
a white background and it kind of as a frame. Now it didn't do exactly as ask, since there's no square
frame around it. So now let's move on to a slightly different prompt
colorful round flower frame in a watercolor style. Now you could do
square flower frame or diamond shaped frame,
I guess would work too. But this worked out well since we're now
getting a round frame. Now to get our
design dimensions, we can head over to print fold, which is a print on
demand marketplace. And download the templates for their 11 ounce mug and a template for their
15 ounce mug as well. Okay. So here we
are on Red Bubble, which is another marketplace. And I want to show you really quickly what happens
when we put these in. So the first one that we did, the full flour, I'm going
to bring that in here. I just want to show you
what it looks like. Let's go down to mugs. When it loads up, we're going
to turn everything back off so you can see it works on some of these
other ones as well. But the mug is down here. All right, so perfect. When we go with 100% on
the slider on the right, 100% size on product,
The mug fills it. And then I could just save it, so you know that's
it for this video. Journey can also be really helpful for print
on demand products. We've only done mugs
in this tutorial, but you can also do that
on shirts and stickers, and also a lot of other
products that are offered as a print
on demand service.
16. Design Characters With Midjourney: This is oil. I'm going to
show you how to create consistent characters
with mid journey. We'll go, we're keeping consistency in different
actions like running, sleeping, reading, and a couple
of others that you might need for characters that
need to remain consistent. Let's get started with our first mid journey
consistent character prompt. Type the imagined command, and start describing
your character. So let's say a little
fox character now. It doesn't have
to be a fox here. You just have to get creative
and choose your animal, cat, dog, frog, anything
that you want, really. Once you've done
that, the next part of the prompt is
even more important. Most important, I would say type in multiple poses
and expressions. This is going to create
a character sheet with multiple characters that
have multiple poses. So let's narrow
down our use case. So let's say I'm going to add children's illustration
book style. And then just add some
characteristics like simple, cute, and full color. By the way, you can get this
prompt from the description details of your character
like yellow shoes, yellow hat, yellow hooded dress. Just add the characteristics and clothing style of
your specific character. Here as an example, I'm using all yellow clothes. Now for the illustration book with consistent character style, I'm going to use
flat color line. I'm going to use flat
color and no up line. That's pretty much it for our first consistent
character prompt. So just press Enter and you can see we don't just
get one character, but four sets of
different characters with multiple poses and expressions that
all look similar. It's important to not
get too excited about your first prompt just
generated multiple times. So this way you get
many more choices. Just continue generating
these character sheets until you see the character
that you really like. This is a bit of
a random process as you may have
figured out by now. But when you see the
character that you like, make sure to get the seed
number of that image. For example, one of the best
generated characters here is this one little
fox character, this character
sheet, it's one of the best ones to get
the seed number. Go to the upper right corner, click on More Add Reaction and click on the
envelope option. As soon as you do that,
mid journey will generate an alert in the
upper left corner. Just go ahead and
click that alert here. Journey is going to give you the seed
number for that image. Right click and copy this
number and save it somewhere. Now you can use
this number later. Now click the image again
and open it up like this. Right click, Copy image. Now I'm going to close
mid journey and open up the images locally and crop
the characters that we get, individual characters
for each image. When you have the
individual image, you can use an
online tool such as remove BG to remove
the background from these characters,
from these images. Once you've done that with all the characters
that you find helpful, you can just copy
the image you have. You can go back to Mid journey. Click on the chat, miss box, and press Enter Control. We paste the image
and press Enter. Now for this story, I'm
just facing three images. The more you paste, the better
the result you will get. Now scroll up to the
first pasted image. Click on the image
to open it up, right click and choose
Copy Image Address. The whole point of facing these images was
to get their URL, so that we can use
the URL in a prompt. So let's start the
new prompt then. Okay, slash, imagine
and paste the URL here. Do this for all
images separately, which in this case is
three different images. So all I have to do
is copy image address again and paste it
for all the URLs. In the prompt, make sure that you separate them by space every time you
paste a new URL. Once we've done that
for each image, then all of these
reference images are going to set
the standard for the mid journey prompt to make our generated character
consistent to these images. Now we can paste that
original prompt. After these image URLs, we've got the prompt
text as well. Edit out those little stars and remove multiple pulses and expressions and
replace that with specific pulses like
running, reading, sleeping. So we're narrowing down now. So let's go with
run and running. And the most
important thing here is that seed number
we copied earlier. Right, so we saved that. And now we're just going
to go ahead and copy and paste it into this prompt. So seed, and then 319-36-9622
And as you can see, not only have we generated a consistent character to our
original reference images, but we've also shown the
character is running. In those images, we have four different versions
of the running character, which was the action we
specified, the running action. There's more things you
can do with this method. Let's say you already
have an image of a particular pose you want
the character to have. You can just paste that
image here in Discord, and it'll try to replicate
the pose of the image. Make sure that the image that you're trying to copy the
pose from is generic, otherwise we'll try to copy the style from that
image as well. So I went ahead and
searched Google for some generic running
images to pater, like you can see
this link here from Freepik.com which is a good
resource for images online, put in space and then paste
the original prompt here. Make sure to remove
the stars and some other stuff
like the brackets. And then just basically use these links as
reference images. Journey is going
to pull the images from that online resource
and press Enter. Let's just see what happens now. Adding the additional running reference image has dramatically improved the movement
of the character. Here, this is a very
powerful method to get consistent
characters in mid journey.
17. Course End: Congratulations on
completing the mid journey, A I image generation course. You've come a long way on this incredible journey into the world of AI
powered creativity. As we wrap up this course, let's reflect on all
you've accomplished. Here are some key takeaways from your mid
journey experience. You've gained a
solid understanding of the mid journey
user interface, making it feel like
second nature. You've mastered
prompt techniques, enabling you to generate images
that are uniquely yours. You've delved into
advanced tricks giving you an edge in the
world of AI creativity. You've completed four
diverse projects, showcasing your new found skills in real world applications. We've heard from
some of you of how Mid journey has
transformed your, how it's been a game changer
in your artistic endeavors. And how it's become
an essential tool in your professional life. But remember, your
journey doesn't end here. Mid journey is a dynamic
and ever evolving tool. Continue to explore, experiment, and push the boundaries of
what you can achieve with you. Ever find yourself facing
challenges or need guidance. Our instructors and
support team are here to help stay connected with
our mid journey community. Share your work, collaborate
with fellow creators, and keep the creative
spirit alive. As we close the chapter
on this course, remember that the skills you've acquired here are
just the beginning. Your creative journey
is boundless, and the possibilities with
mid journey are limitless. For those eager to
explore even further, consider our advanced
courses where you'll dive deeper into the
world of AI creativity. Including running open
source AI software such as stable diffusion
and sono bark, and other exciting AI tools
that are popping up each day.
The B., Tech Made Easy!
