Transcripts
1. Image Illustration with AI Intro: Welcome back, guys.
The digital skill set that we are
going to be talking about in this class is image illustration with
artificial intelligent tools. We are also going to discuss
the latest update in AI platforms such as Mid
Journey and Blue Below. Additionally, we are
going to be discussing some important AI image illustration terminology
such as image prompt, image set, xpect ratio, image upscaling, image presets are moded as well as
image variations. We are also going to be
seeing how to create various kinds of
images in this class. Additionally, and
most importantly, we are going to be learning
the new method to create a consistent character
image without using sets. So, guys, I advise you to practice along while
this video is going on so that you are going to be learning easily and quickly. So guys t tuning and see in the next video.
Thanks for watching.
2. Latest Update on Midjourney and Bluewillow: Video, we are going
to be talking about some updates in the mid journey and the blue willow platform. So guys stay tuned and identify the new updates in
these platforms. See you guys in the next video. So, guys, in our
previous classes on image illustration,
where we enter, either the Blue Willow or
the Mid journey platform, we asked to join
the free Better to have access to their
discod channel. And when we join
the free Better, we sign in with our
discod account, and we are directed to their discod channel
where we will be able to create our images using the
slash Imagine and command. But now things have changed. It's no longer that way. There have been changes and updates in these two platforms. So we are going to find them out in this video.
Stay tuned, guys. So guys, as usual, we enter in our browser mid journey.com to be able to access the Mid
journey interface. Now, when the website opens, we see an interface like this, as usual, will be asked to
either login or signup. If we don't have an account
in discode account, we can click on
Signup to create one. But if you are with discod
account you can click on Login and continue login
with our discode account. We continue with discode
by clicking on it. Now we allow for
time to get access. We allow for time to get access. Now we click on
Authorize as usual. Now we've done that, I will now direct us
to the Mid journey, the new Mid journey platform. As you can see, there is
observable changes in the Mid Journey platform unlike how it used to be previously. Now, the organizer has to do with the previous
works you've done. You will see all the
works you've done or the images you've created with
the Mid Journey platform, you will see it in
your organized stab. Now, your account is
at the bottom left. Monday subscription,
like I said, Mid Journey is no more free. There is no more free
version of Mid Journey, so you have to
subscribe to one of their plans to be able
to access the platform. So when you click on
the bottom left icon, you click on Mondays Upload, you are going to
see all the uploads you've heard on the platform. You click on Showcase, you see the communications and where
you enter your prompt. As you can see, creation is being tested with subscribers. So if you're not a subscriber, you'll be having access to create content with
media assistance now on maybe a rescission is being lifted
subsequently, which I doubt. So this is the community
creations where you can just have or get a prompt idea to creating a specific
type of image. Now when you go to
the chat section, the chat session and click
on the general Kios, you are going to
see it's a room for general proposed creation,
discussion, and chaos. This is one of the
rooms available in the new mid journey platform. Now, the prompt craft is where
you can get a prompt idea. You work with other
users to craft the perfect prompts or images
for your projects or works. So as you can see,
there are many of them. Newbies, if you are new
to Mid journey platform, you can start creating
images from this section. So if you're newbie, you can just start getting inspiration from other creators. Now, the light mode is to change the mode of your
screen as usual. So that is it. So there are observable changes in the
new mid Journey platform. It's now mob free
like it used to be. And before you can
start creating content, you have to desuscribe. Let's go to Blue Willow and see also the new changes in
the Blue Willow platform. As usual, we enter
in our browser blue willow.ai to be able to
access the platform. Glow.ai in our browser. When we've done that, an
interface open like this. As you can see before, there is no space for us
to start entering our prompts immediately from the interface from the
platform that opens. But now we can do that
because they have provided a prompt window for us to
start entering our prompt. Now, we are going to
do that in two ways. It's either we start creating
it immediately from here, as you can see, by
entering a prompt. First of all, let's try
entering prompt and see what M journey
would do next. We want to create an image of a golden lion wearing a crown. Remember, this is for
demonstrative purposes. So we are going to
create that image in Blue Willow and see how
it's going to turn out. This will help us observe the
changes, the new changes. So we click on Genet Work. And when we've done
that, as you can see, it will now direct
us to Lime wire. I will send this in details. If you have been
using Mid journey, if you have been
using Blue Willow, you notice that this is
not how it used to be. For example, when we enter a discord after logging into
the Blue Willow accounts, we'll be entering or assessing their discode channel
with our discode account. And when we do that, an interface we open that is similar to the old mid
journey interface, we have access to different rooms to join
and create images, as you can see with the
slash Imagent command. But it is no more that even if you go
through your discord, you'll be asked to join Lime
wire. Welcome to Lime wire. You can see in the screen,
creates AI images and videos. So that is the
observable updates and differences in
these two platform. We can no more join
rooms and start creating our images with the
slash emerging command directly from this room anymore. No, we are going to be directed to me wire, the Lime
wire interface. Now, Blue Willow was
acquired by Lime wire. Lime wire is a platform that
has been around for ages. It's kind of a block
chain platform. So it's introducing or incorpating block chain
technology into Blue Willow. As you can see, the home
features in our rights. It is where we can
see the features. The explore, we can see
is like a community of creations from other community
members using Blue Willow, where we can have a prompt idea for any image you
want to create. Now, the library tab
is the record of all the images
we've created using the new LimeWire Blue
Willow integration. Subscription. Unlike me, Johnny, Blue Willow C
offers a free plan, so it can create images,
but there are restrictions. So if you want to remove
the restrictions, you have to upgrade to
a professional plan. You have to be a subscriber, so you check the plan that
best suits you and upgrade. Now, the payment, like I said, Blue Willow integrated with our black Wire two
cover Blue Willow, and they are introducing
their blockchain system, they're introducing their
blockchain technology into this new platform. So as you can see,
you have access to deposit in your dita wallets, make deposits and withdrawals, buy their token, the
LMWTken and so on. If you're familiar with NFTs and Blockchain, you
understand better. Now, my page is
your profile page. Everything about
you, your images, your account setting will be on your profile Blade
The settings page, I edit your credentials, you can click on the settings
to edit your credentials, have access to API keys, integrate your social media
accounts and the rest. Now, the partner program
has to do with commission, like an affiliate kind of tin. You enter into a partner
program with them. You collect your affiliate link, you share your affiliate link, and whoever joins, by whoever
joins to your partner link, you'll be having a
commission on that. So I'll be doing a
more detailed video on this lime wire specifically or a separate class for that. So guys, watch out
for it to be able to understand this block chain
technology integration of lime wire with blue Willow. As you can see, the lion with a groding crown we created
initially has appeared. So as usual, you can
see the download icon, can see more like the
variation upskalin image or painting image like we used to do in the
old Blue Willow. So click on the edit, if you want to enhance our prompts and create more images or a
new image, click on edits. Now, if we want to
create many variations, create variant to get different
variants of our image. Can upscale, we can
move background. We can do image music,
and we can publish. When we click on publish,
it automatically appears in the public domain. And whoever, as I write, whoever assesses whoever views our images, we add some tokens. Remember, I said I'll be
doing a detailed video on Blue Willow integration,
so watch out for that. On interface, you can also
have access to other tools, by clicking on AI tools. You can see other functions of lime wire from that interface. You also see functions
by clicking on more. And when we click on
the right top right, we see our token, want
to buy more token, get access to our wallets. We see it there. Now, the
models are still available. We have stable diffusion,
we have Blue Willow, and other models for
pro subscribers. You know what model list,
we'll be talking about that in the Mode one of this
class, so watch out for that. Discussing aspect ratio,
talking about SID, talking about
variation of images, upscaling of images,
and mode image models. So I stay tuned in this class, you'll be learning what those
terminologies is all about. So that is the blue
willow interface right now taken
over by lime wire. So we can also assess
other functions in our screen to get used
to their platform, the new platform assess other functions as
consequent create image, edit image, and paint
image, upscale image. So we go back to
home by clicking on the home tab, when
you click on it, we ask us to start our fresh, either by assessing
the assistant, the image or the
audio music function. So that is the new
feature of Blue Willow. The updates, as you can see, is mind blowing is so
wonderful and fabulous. We'll be learning how to
use that lime wire to the flest for monetization purposes and for content
generation purposes. So you guys stay tuned in our subsequent class while we talk about Blue Willow and Lime
wire integration in full. See you guys in the
next video. Thank you.
3. Module 1 Explaining AI Image Illustration Terms: Guide to the Mode
one of this class. Before we start creating AI images, we have
to first of all, understand terminologies associated with AI
image creation, and that is where
prompts come in. We have the expect ratios,
we have the models, we have the upscaling, we have the variation,
we have the sets. We are going to understand
this terminology so that creating images with AI
will be very easier for us. So stay tuned and see
you in the next video, we will be explaining
them with demonstrations. Welcome back, guys. So
the first terminology we are going to be
explaining is prompts. Now, what are image prompts? Prompts are descriptions
about our image. Taking into cognizant four
aspects, the objects, our image model, the action
our image is undergoing, and the attributes of our image. So when we go all
these things together, we are going to be
able to develop a command that describes
our intended image. And we can do that by typing it in any of the image
generation platform, typing it or describing
our prompts naturally. If we are good in analyzing
or in description, we can type it in naturally. And when we do that, we click on Enter and whatsoever platform
we are working with, and it generates our image based on the prompt
description we gave it. Taking into cognizance,
those areas I mentioned, the model of our image, the object inquest, if
it's a male or female, the model has to do with if it's a three D or a realistic image, then expect ratio, if
it is, if need be, then the action, what is our object doing in our
image and the attribute, what our object is looking like? So when we take into
cognizance these areas, we developed our prompt
and the platform will generate our image
based on the description. Now, if we're good
in storytelling, we can create it naturally. We can describe it ourselves. But if you are not
good in doing that, we have another way
we can generate our image prompts and that is
with the help of Charge PT. So when we go to our
ChargePT platform, as you can see, integrating this IPROM extension
helps generate prompts. As you can see there
is a mid journey prom generator day already. So we can use it to
create prompt easily. When we enter the command. We enter the command like create a pizza style generative art
prompt of a young cowboy. So we have given it an idea of the image you want to create. So Charity bit will
help us describe the scenario, describe
the attributes, describe the model for us in a more simple but
complex manner. So instead of
cracking our brain, trying to formulate a prompt, you can just enter
a simple command and allow ChRGPT generates a full detailed prompt for a
high quality image for us. As you can see, it has started generating
our image prompt. And when we copy this
team now and enter into whatsoever platform we are using for our image generation, it will give us a
well detailed image without leaving anything behind. These are two ways we can create image prompts
for our work. It's either by going
to ChargePT to enter a few details and ask you to generate an image prompt for us, or we can describe it manually inside whatever
platform we are using. Inside the prompt bar of whatever image generation
platform we are using. Is it me Johnny?
Is it Blue Wheel? Is it Lonado AI,
Microsoft copilot? Is it Dali? Anyone we are using. So the next one
is aspect ratios. Aspect ratio has to do with the dimension of
our image frame, the size of our image. Would it be a square image? Is it a landscape image or
is it a portrait image? Now, depending on what you
want to use our image for, is it for YouTube video, Aisin our image for
Facebook advert, Asimge for TikTok AAT? As you can observe,
those platform have different image frame rates. So that is what we take into cognizance when we are choosing our expect ratio for
image generation. As you can see, each of the
image generation platform have a section for image dimensions where we can click on them after
entering our prompt. Now the next thing is our model. For example, we are
using the Leonardo platform to explain the model. Model has to do with
our image style. Is it a cartoon style
image you want to create? Is it a pixel star image, a three D star image? Is it a realistic image? Is it a cinematic
image, a dynamic image? So we have to select
the image prompt to be able to general the intended
image model we have in mind. So we'll go to the
model section of any of the image
generation platform, we select the one we
are working with. When we enter our
prompts, First of all, we enter our prompts in the prompt bar of the
image generation platform, we select our image model, we select the expect ratio. Once we've done that, make the necessary
negative prompt edit and we click on generates. So that is how we work with
all these three aspect. Now, but what about image variation and image
upscaling and, you know, SID? What about those ones? Now,
when we've created an image, we want to create we selected
a specific type of image. We want to work with, and we want to create variations of that
particular style. You know, when this platform
generates image for us, they give us four image at once. So when we pick one out of
the ones they've given us, I want to create a more variation
of that one we created. We can click on
create variations, and that selected when we have
multiple variations of it. Now, what of upscaling? If we want to have a high quality resolution
of an image created, we can click on the function
of upscaling in any of the image generation platform to get a more high quality image. Of that platform. Now
the next one is SID. SID is like a unique identifier, a unique number given to any of the images we've created
for reference purposes. In case you want to use that image in a
subsequent image iation, we can make reference to
that image using the SID, which is a unique
number attached to it. If you go our video, our class on content writing
and image illustration, we did previously, you
can see how we use SIDs to generate consistent
character images. So, guys, these are some of the terminologies that is being adopted in image generation
with artificial intelligence. So I stated where we
demonstrate how we can use this temolos in derating images in different platforms
in this class. See you guys in the next
video. Thanks for watching.
4. Module 2 Introduction to Dall E: So in this video, we are going
to be introducing one of the AI tools we use for
image illustration. Practice along while this v is going on, seeing the next video. Guys, the first AI two
we are going to be introducing in digdo
is called Dali. Dali is an AI AI image generation
to develop by Open AI, the makers of CharPT. So in our browser, we enter DL E. And when we do DL we
allow the interface to open. When it opens, we see you have to select
the one by Open AI, DL two, we select it. So now we are going to be asked to either
login or try ChRGPT, but we are not trying CHAPT, so we are going to still try DL E. When we click on Duct Tab, the interface will open and
we'll be axed to continue. Remember, if you are a
subscriber to ChargePTPlus, you won't have problem
using this as. It is not a free version. It's not a free version. So when we click continue, we will be asked to buy credits
for $15, as you can see. So if you have the money, you can just buy the
credit and you'll be given access to start creating
images from the DLI platform. So but there are
ways you can try this Dali Image
generation two for free, and I'll be demonstrating
that in the next video. As you can see, we can't enter our prompt because
we have no credits. For us to do that, we have
to purchase a credit, which will allow us assess to start entering our prompt and
generating various images. So as you can see,
we can't do that. It's not at the command, Is not allowing us access. Even if we've
entered our prompts, for example, I want to enter a prompt now to
demonstrate for you. Entering the prompt,
I want to generate an image of a young African
boy riding a horse. So remember this is for
demonstrative purposes. So I entered the prompt, a black African boy riding
a horse in a green field. And as you can see, I can't generate image because you're out of credit,
tobia out of credit. So for me to generate any image, I'll be purchasing credits
to allow me do that. So when you want to buy credit, for example, you can
click on By credit also. I can see image samples
of community creations, community members
creations, as you can see, just like in every other
platform I've shown you. Now, when you click
on By credit, you'll be asked the
amount you want. If you want to buy 230 credits, just according to preference. So when you've selected
your preference, all you have to do is to
add your billing method, and once you've done that, you click on Continue. So that is it with
using this Dali. It's not free. You have to buy credits for you to
be able to use it. But in the next video, I'll be introducing to you the free version of
Dali where you can just write images with Dali by
opening your eye for free. So stay tuned and see in the next video.
Thanks for watching.
5. Module 3 Image Generation with Dall E : So, guys, in our browser, we enter Dal E free. Make sure after typing
Dali, you type free. And when you do that,
you are going to see Dali free as an option in
the search engine results. So you select it, as you can see from the screen. Now, you've been granted
access to use DLI for free, and you are going
to be generating images straight from
the prompts bar. So as you can see, these
are community creations. You can also go through
the whole website to know to get more information
on how it works. And if you have any other
information to grabs, you can also do that sign in if you don't have an account, pricing, as you can
see, are pricing plan, you have the zero, you
have the free version. You have $5 per month. You have $10 per month. So for the purpose
of this video, we are going to be
using the free. So we enter our details. We sign up as we can
save our sign up details and for subsequent uses in case if flocked
out of the platform. So we go back to home. Now,
we see the communication. We want to create our
own images straight. As you can see, we've
already created accounts and it's given access to start
creating our images. So the first thing
we have to do, like I said, we
enter our prompt. I've told you want to prompt is an image prompt has to do with the description of your image. Taking into cognizance aspect
like objects in question, the attributes, the model, you want it to come out and the action your
image is performing. So that is what you have to take into cognizance when
generating a prompt. Now, as you can see, we
are typing it naturally. If you don't have
any prompts of hard, you can go to CharPT and ask you to generate the prompt
for us and paste it there. So once you've
entered our prompts, I click on generates African
boy putting on a cowboy hat, riding the black house
in the greenfield. As you can see, they generated
an image for me free. Here is my generation, and they've generated an
image. So this is free. I can just right click and save my image directly to my
image folder and my device. So that is how you
generate image easily. With the leafy. Now, depending on the intended
purpose of the image, remember your prompt
determines that. Whatever image you
want to generate, your prompt determines that. So you make sure your prompt describes your image in details. Like I said earlier,
if you're not good in description
or storytelling, you can always use ChargePT
to generate a prompt for you. You can also make edits like
I've edited this prompt, and I changed it from
a hart to a face cap, and you can see it
has given the boy my image description a face cap instead of a hart,
as you can see. So this is how you
create images. Go to the pricing, check
your restrictions. You can go on with the
free pasion, fine. If you can't go with
the pre version, then you can subscribe to
any of your paid plans and start using it for creating
your professional images. Remember, Di is a very high or highly advanced
AI image generation tool because developers open AI
the makers of Charts JPT. So you will not run short of any image quality when you are generating
your images with Dali. So, guys, we are going to be introducing the next AI
tools for image generation. So I stay tuned and see you in the next video.
Thanks for watching.
6. Module 4 Introduction to Leonardo AI: So the next AI image
generation to we are going to be introducing is
called Leonardo AI. It's also one of
the good AI image generation platforms
we have out there. So in our browser, we type in leonardo.ai and allow
the interface to open. When it does, you are going to see an interface like this. Now we can go through
it to understand more about their platform. Now, we click on features, we
see the features solutions, we see the solution
they have to offer. Interior design, graphic design, print on demand, architecture, photography, and marketing tools are among the
solutions they offer. Now, the resources,
as you can see, go through the resources
to know what it entails, what they have in store for us. That is all about the resources. The teams, if you have a team you're working
with or if a developer, you can also go to the team
and developer section. When you click on Sign Up, we enter our credentials. You can use our email or we can use Apple ID or we
can use Google. So for the people that did
video I'll be logging with my Google account and
I clicked on Google. So I selected an account, one of my Google accounts. And when I do that, it will automatically synchronize
with Leonardo AI platform. As you can see, it has
granted me access to the Leonardo AI image platform. As you can see, look at the features like
realtime Canvas, real time diation, motion, image creation, upscale,
Canvas editor, and more. Now, if you scroll
down, you are going to see some of their new tools. Like style preference,
content preference. Now, if you go down a bit, we'll see some of the
community creations. So like I said earlier, you can generate
your prompt ideas from community creations by clicking on the images and copying their prompt
to make some edits. The left hand side
is our accounts. We see our accounts.
See the settings, if you want to log out, that is where we are going to
be logging out from. As you can see, it's
partially free. So look at the tokens
they've given us. We have one and 25 free credits. So if we want to
upgrade, we can upgrade. Our personal feed, when we
go to our personal feed, it has to do with our
image creation history. He will see the history
of all the images we've created with the
Lonardo AI platform. You can see the follower feed if our followers on our
accounts like feed, if our images are being liked by other members on our collection, so we are going to assess
it from this section. Now, if you go to the
image creation tab, but this has to do with creating
our image from scratch. How by entering our prompts, selecting our model
style aspect ratio and negative prompts and
clicking on generate. Now, remember, we have
125 free credits, and as you can see, these are some of the
images we've created. In the left hand side,
you see some functions. When we click on that Pres, you see the models
available for us to use. You have the cinematic model, we have the concepts
and all other models. Now, Auto can see the prompt enhancer you
can see a preset style. Like I said, I selected
a press style of my choice for my model like the three D render is
being selected there. So whatever images I'm
generating will come out in a three D render form. That is all about models, like I explained in the
terminologies section. So contrast, you can
decide to choose your contract if you
want a high contract or low contrast or
medium contracts depending on how you want
your color to appear. Remember, the generation mode, I'm not a subscriber
in this plan. So if you hire a subscriber, having access to
that quality tab. The image dimensions has to
do with the aspect ratios. What images are you creating? What is the frame
rate you're using for whichever platform
you want to post it so it will determine the image dimension you
are going to select, and that is the aspect ratio. So the number of
images automatically for our credits and
for our free plan, it will be generating for
every prompt we enter, it will be giving us four
images so as you can see. Now, private mode feature
is for subscribers. So if you don't want
your picture to appear in the public domain,
you can switch it off. That is for private mode.
Now, negative prompt section, if things you don't want
to appear in your image, you can select it in
the negative prompt. Then transparency is there. Telling is there. So depending on whatever function
you want to use, then look at the fixed Sid. Like I said, Cs are
unique identifier. So if you want to give an image, a unique identifying number,
you can enter it there. That is all about the
advanced settings. Remember the negative prompts, anything you don't appear
in your image can be stated clearly by clicking on that negative
prompt and Anthony, such as you can
say you don't want shaky hands, blurry vision, or whatever negative
prompt you don't want, enter your prompt. And when you do that,
you click on Generate. So that is how you assess
this image creation too. When we are going to be
creating our images, you are going to see how we
are going to be doing that. So you can stay tuned when we
are going to be doing that. Like if you enter our prompt, whatever prompts
you enter there, we are going to be
generating image with it. So remember, use
our E model select our expect ratio before
generating our image. And when we go back
to the left tab, we see other functions
of this Leonardo AI, like the motion function. What our motion function where we want to turn
an image to a video, we are going to be
demonstrating that subsequently in our next class. So stay tuned and
see how we are going to do that in our
subsequent classes. Real time Canvas, if
you want to start from scratch to generate your
ideas through sketches, when you sketch, Leonardo, I will give you a
picture demonstration of that sketch you've made. As you can see, I'm sketching, and it is giving me
ideas from my sketch. Is generating images from
my sketch, as you can see. So it depends on me now to start making some
necessary edits. That's if it got
my idea correctly. Now, in paint, if you want to scribble on your already
generated a canvas, you can do it there. And if you go to
the left hand side, you see some other tools
like eraser, brushes, color. So that is what is available on the left hand side at
the inpaint section. As you can see, you can increase the intensity of your
brush, and that is it. So if you want to add image, remember, you can see your
uploads in that section. I want to add prom to
enhance the images, you can click on that tab to add the prom that to enhance the
image or you dig the image. That is the brush icon if you
want to inpaint and rest. Let your brush size,
select the color, and it's going to be
how it turned out, or hives, set it as you are selecting or making the
changes in that section. So that is a can see. I change
the brush mark to a bold. Painting, so as you can see, this is giving me that a. So that is the undo. If you want to go back to
your previous designs, you click on that Mac or
icon at the bottom left. So we exit. You go
over to the next two, that is the Canvas editor, you have the real time Canvas. You have the real
time generation. So what is real time generation? As you're entering your prompt, once you enter your prompt, it'll be directing
it automatically from the beginning to
the end of your prompt. So once you enter your prompt and the image
is being generated, if it's not what you want, you can easily edit it immediately. That's real time generation. The Canvas editor,
if you're working on a project that has to
do with various images, you can group them
under one canvas, and that will be
very much easier for you to work on the project. That's for the Canvas editor. Now, Universal
Oscar, like I said, an upscalar is
kind of increasing the quality of an image
that is being generated. For example, when an
image is being generated, four images being
generated for you, you can decide to upscale. Now, if it's not your image, you can also import an
image source it from your device folder and
start working with or you source it from the
image enertd on the Lonardo AI platform in your product or
your image section. That is what you use
Universal Upscalar for to upgrade your
image quality. You have other features. We have the fine tune models, training of data sets,
texture generation. So as you can see, there are so many features
in this Leonardo, like the fine tune models. You can see samples of models, unado AI Platmhart and how their seats or
their prompts are. So you can also get ideas from
the Training and data set, you can train your own model, train your own Avatar, and if you're into comics,
you understand better. Train your own
dataset for you to use it consistently
in your project. Texture generation. You want to have a three d
texture generation, you can also use it
from that angle. Remember that is for
paid subscribers, so you can use that very
well if I paid subscriber. Being a paid subscriber opens or grants you access
to many features, advanced features in
this Lonado AI platform. Like I said, this
platform is one of the top notch AI image generation platform
we have out there. So, guys, if you want to work professionally
with an image generator, you can also use Londo AI. We have other access tools
like the API Assets. You want to integrate
it to your website. And you have the Setans has
to do with your details. You want to edit your details and your preferences
and your interest, you cannot do it from
the Setans side. Remember, API to Deta be
integrated to your website. Now, if you want
to write an image, you can either click on
the image generation in the front screen or
you can click on that image creation tab on the left hand
side of the screen. Remember, we are going to be generating images with
this Leonardo AI platform. So stay tuned and see in the next video.
Thanks for watching.
7. Module 5 Image Generation with Leonardo AI: So now we are going to be
generating our images with the Leonardo AI image
generation platform. So guys stay tuned and see
how we are going to do that. Well, come back, guys. So in the Leonardo platform, we're clicking on the
Get Started button. And when we do that, we
sign up or sign in with our Google account since we've created our
account with it. So in the interface,
we are going to be seeing some functions and
we want to create an image, so we click on the
image creation button. And once we've done that, it will open an interface
for us to start creating our image or start
generating our images. That is the prompt bar
where will be entering our prompt by typing it in or
by copying it from CharPT. So remember, like I said, if you're not typing
it naturally, you can copy it you can allow ChargP generate it for you
and you copy it from there. So look at the prompt we generated when we are describing what prompt terminology is. Now, we are going to be
copying this prompt, and we'll be pasting it in our Leonardo AI platform
prompt bar, so we copy. Can highlight to copy
or you can select the copy icon at
the bottom there. So we highlight, we
copy it and we go back to our Leonardo AI
platform and paste our prompts. And once we paste our
prompts, we edit it, we go through it to make
sure it aligns with the image you want to create or the image you
want to generate. And now we start
setting the parameters. We click on prompt enhance, you can decide to
switch it on or off depending on
your intended use, or you can decide to
leave it automatic. So we left it at Auto. Now we select our preset style,
that's a three d render. We go to the preset,
leave it at Phonics. There are many other ones, but
we leave it at the phonics because it's kind of giving
us a three d render formula, so we leave it at Phonics,
Leonardo Phonics. So we've selected
our preset models. The next thing we are
going to be doing is to select our expect ratio, that's the image dimension. So the quality aspect is for
subscriber, so we leave it. Now, depending on intended
use of our image, we select the frame rate
and we are selecting the system by nine because
we want to use this image, maybe for YouTube video
of a Facebook square. So we select a system by nine instead of a one
is the one ratio, we select system by nine. Now, the number of images
you'll be generating is four because we can't change that since we are
using a free plan. So the private mode can be switched on since
we are using a free plan, and we can't turn that off or we can't turn that on actually
because the plan is free. Going to the advance settings, we click on the negative prompt. Like, the negative prompt is all about some functions we don't want to appear in our image or some corrections
we want to avoid. For example, I've entered some negative prompts
I don't want like blo shaky or
irregular body parts. So Leonardo AI will not be generating image
that has a blur, shaky or irregular body part. So that is it for
negative proms. So once we are okay
with our prompt and the parameters, we
now click on generate. And once we've done that,
we patiently wait for the Leonardo AI platform
to generate our image. As you can see, it has
generated four images. Remember, we can reduce
the number of image it generates by clicking on or selecting the
number of images. But since we are using the
free plan, we can't do that. So as you can see,
these are the images. So we go through them and
look for the one that best suits our intended use
we click on the image, we'll see some functions
like the download button, the image zooming button
for us to zoom image. Remember, it will
cost us some token since we are using a free token. Then generate motion video, we can giret videos directly from our images from this stub. Now Oscar, we can oscul our images or also we
can edit it in Cat. Unfortunately we can't delete
those images we don't want because we are
using the free plan and we are not the subscriber. So now we go to the images and select one that we want to use. Remember, we can
also edit our models by going back to
the model section or preset session and
select the style we want. So we change it. We selected a portrait cinematic style and let us see how
it will turn out. So remember, we are using
one particular prompt, but we are changing the outcome. We are changing the model look. So now we've done that,
and we click on Generate. So we patiently wait for
the platform to generate based on the new selected
model style we picked. Now, we patiently wait while the platform is generating another model style
for us. So that is it. These are the previous
images we created before. So we can click on any of them
and perform any function, any further function you want to perform, or we click on it, you can either
download directly, we can either upscale it
or we can either create variations of it
from this point. So once we are done, we look on the icon to cancel the tab and go back to
our previous window. So our image has been generated based on the
new model we selected. As you can see, the scenario has changed because we
choose another model. Instead of the previous one
that was three D render, we selected another one, and since we've changed the model, look at how the
scenario has changed. But one thing you have
to bear in mind is the cartoon niche
look is still there. The cartoonish look is still the body scenario has
changed. So why is it that? Because in our prompt, we left it at animated. Within it. I did
our prompt folder. We did move the
animated features, and that is why we
are still having a cartoon like image
being generated, but different scenario, different model because
we've changed the model. So when we click on
any one of them, can decide to perform
any folder function, if that's our selected image, decide to download it from there or upscale it or
create variations. But how can we change our object from this
cartoonish look? Even if we change our
model? How can we do that? So we can simply go over to our model to select the
intended look we want, try to change it to any of them. Once we've done that,
we can decide to now edit our prompts, also. So just watch how I'm
going to do that. So we want to change our
intended object look. So we go back to our prompts. As you can see, it's still
in that pixel style. Pizza style has to do with a three d render kind of style. So we are going to edit it
into a realistic style. As you can see, I change
it to realistic style. Now, I'm going to look for areas where animated is
being mentioned, and I am also going to
edit that to realistic. So I edit animate it entirely, or I replace it with a
realistic explanation. So as you can see, I have edited that and I
have deleted it. So I go through my
prompt again to look for areas where
animated is being mentioned. And once I come across that, I'm going to remove it entirely, removing it entirely and
leaving it or replacing it with cinematic style, just
like I have done here. So once I'm very sure that
there is no more mention of animated or cartoon like
mentioned in my prompt, I will now go I
will now go back to my model and select
the model I want. I'm selecting I want to create a realistic image,
a real life image. So I'm going to pick
the cinematic model or a cinematic kind
of real life look. Once I've done that, the
precess has changed, my precessor have changed. I now click on Generate. So guys, watch out
and see the image we generated from this
edits how to look that. Remember, we now have only
100 tokens left because we are using a free
version of Leonardo AI. So the more image we create, the more our tokens is
being reduced or deducted. So as you can see, the
image has been generated, and we have now a
realistic look kind of real life image
of our prompts. So no more a cartoonish look
or a three D kind of render. The background is real, and the image is a real life person. So that is a realistic image. So when you are hearing, want
to create a realistic image means image resembling
an individual person. Or a real life scenario. So this is a realistic image we've created from
our initial prompt. So we can always use
one prompt to play around it to generate different styles and
different models of images until we
get the intended one we want to use or
the intended image that serves the purpose
we are using it for. This is our prompt. Now,
if you want to change some artery boots of the
object like as cans, we want to change how
the objects look. The man that is being generated,
the man in our image, want to change how he looks or want to change
some of his artery boots. Now we can edit our prompt, leaving the main objects. As you can see, we remove
other artery boots, but we left the main object. That is the young cowboy standing in a vibrant
desert landscape. Now, we want to change
the artery boot. So I said he's putting
on a red boot, black hart and
maybe other things you want to add to the prompt to change his At boots or
his looks or his actions, you can also do that
further in your prompt. So once you've done that, and you are satisfied with
the prompts you entered, you can just click congenerate. So look at what I've done. Returning the objects
of my prompt, I now change the
Attributs and action. So let's see I
click con generate. Remember, I left the Xpet ratio and change I left the
style to cinematic. And once I've done that,
I click congenerate. So as you can see, it has
changed the look of my objects. This is not the clothes
he was putting on before or the boots he
was putting on before. So it has changed. Now, even though he didn't
give us the exact description, we can always play around
our prompt till we get the intended what the
intended style we want. So there's no limit in as
much as we are a subscriber, there's no limits to
the styles or the edit, number of edit you can
make to our prompt. So once we are satisfied, we can either download
upscale or create variations of each of the images or either
of the images we picked. Now, remember, your prompt
matters, from your prompt, your preset style matters, your style matters, your
aspect ratio matters too. So all the matters
in image generation. Remember, so all these
things matters in image generation and don't forget the negative
prompt aspect. It helps to also create or it helps in creating
high quality images. It helps AI tools to
create high quality image. So you have less time
working in your or switching your prompt to get
that exact image you want. Remember, when you
click on any of the image being gireted, we can always edit or download. So, guys, this is one of the tools we are
introducing in this video. In the subsequent videos, we are going to see other
AI tools we can use for creating high quality image. So, guys stay tuned and
see you in the next video. Thanks for watching.
8. Module 6 Introduction to Microsoft Copilot: So another good AI tool
that we are going to be introducing in this
video is called copilot. So guys stay tuned
where we introduced the Microsoft copilot A
two for image generation. See you guys in the next video. So, guys, in our browser, we are going to be entering
the copilot search, and so we are going to enter
copilot.microsoft.com. That is what we are going to
be entering in our browser. So once we've done that, an interface will
open like this. This is the copilot by Microsoft
AI image generation too. And we are going to see
how we are going to use this Microsoft copilot
to generate images. As you can see, there
are various presets with different functions. So I'm scrolling over as I'm scrolling some
changing the slides, you can see various presets
by Microsoft copilot. So what or whichever best suits our intended purpose,
we can select on it. Now, if you go to
the upper section, that chart section or
the Hamburger icon, sorry, I where we change the appearance or the
mode of our screen. Then the setting is where
we change the language, the country, our location, and other features, too. Now, this is Hamburger icon. I've told you that's
where we access that. The charts icon has to do with the history of all our charts. Then the sign icon is where we sign in with our
Microsoft account. So you are going to be signing with your personal
Microsoft account. If you don't have a
Microsoft account, you can proceed in
creating one for yourself. Now when you've done
that, enter your email, Microsoft will send you a code. Now, once you've gotten
the code and entered it, then you click on sign in. So once you've signed in, and you'll be able to now assess the co pilot platform for
your various function. So you select the function if you want to stay
signed or not. And once you've done that,
it will take you straight to the Microsoft interface
for your creations. As you can see, the
right hand side, where you have chart
is where you have the history of all the
works you've done. Now, the plugins
are the integration that is in the co
pilot ecosystem. So that's the integrations
where you have the plugins. So you see your integrations. The Hamburger icon
is for Settings, and that is for your account. So that is it. Now, we are going to be creating an image with
this Microsoft too. Remember, that is where
we enter our prompt. And if you want to
upload an image, you can click on
that icon to add an image to edit
them or enhance it. So that is the share icon, if you want to
export our images, and that is the voice assistant. We can try copilot P. There's
a pro version of co pilot, but for the purpose
of this video, we'll be using the free version to create our images as they
serve the same purpose. So, guys, stay tuned and
watch how we are going to do that in our
subsequent video. That new chart icon
is for you to start a fresh page for your
image generation. We want to edit your previous
page and start afresh. You click on that new chart tab. So guys stay tuned and see
how we are going to create our images in the next
video. See you guys.
9. Module 7 Image Generation with Microsoft Copilot: So, guys, in the co
pilot interface, the first thing we
are going to be doing is to enter our prompt. So we go about to check GPT and copy the prompt that
has been generated from us. Now, once we've done
that, we now go to our prompt buying co
pilot and paste it. So once you've pasted it,
that is how it will be. Now we click on
submit by clicking on that arrow at
the bottom right. So once we've done that,
we now patiently wait for copilot AI engine to describe our image and
be able to generate it. After understanding
our image concept, it will be generating our image. So we patiently wait
for it to do that. Remember, we copied
our prompt from ChargePT and we paste
it in the prompt war. Also, that is it. So we can decide to So, guys, we patiently wait for
our image to be generated. Now, our image has been
generated, as you can see, so we can click on each
of them to be able to download or share or customize for using
the Microsoft designer. So remember that arrow, we can use it to scroll
from that tab or go back to our images to pick
them from the previous window. Remember, we can always use this arrow in our
screen to scroll, but we don't want to do that
here. So these are image. We go through each and every
one of them and select the one that best suits
our intended pupils. As you can see, all
of them look alike, but they have slight
differences in appearance. So we select them individually and go through either we download or we share
or we customize further. Remember, we can
always scroll using that arrow beside the image, the triangle arrow
beside the image. You can always use it scroll. Now this is our prompts. We want to make necessary
edit, remember, we can always edit our prompts
and change their style. So we want to create a
realistic style of our image. So as usual, like we
did in Leonardo AI, we are removing all traces of
cartoonish description and replacing them with realistic
or cinematic description. So in the case of this video, we replaced them with
realistic description. So let us see how it's
going to turn out. We click on submit and
patiently wait for copilot to generate our
image. So we are doing that. We are patiently waiting while Copalos is generating our image. As you can see, it has generated another version of our image. It has generated another
version of our image. For us. Look at this
is a real life person. But the background, remember
within which the background. We left it in a kind of cinematic or three D
render background style. But our object, that is
the person in question, we edited it from a three D render to a
cinematic or realistic render, and that is why the object alone changed to a
real life person, but in a kind of three D
setting, as you can see. So depending on
the intended use, like I said, you can always
create images based on that. If you want to start
on a fresh pad, click on new topic, when
we click on new topic, it takes us back to the
homepage, as you can see, the left hand side
where we have chart is the history of all
the work we've done. So you can click
on it to always go back to our previous image, select on any of the
history we want, and it takes us back to the previous window or to
the previous generations. But we want to
create a new image, so we click on new topic, and we're entering a new
prompt for a new image. Now, we don't want
to use chargeability to generate our image prompt. We want to see how we can enter an image prompt naturally. Based on our capability, our ability to
describe scenarios. So we want to describe a
brief scenario of our image, and we want copilot to generate our image based on the information
we've entered. So we are asking copilot to create a realistic image of
an ancient Indian queen, putting on a red or a blue or a pink or whatever attributes you want to
ascribe to the image, you can always do
that in your prompt. Remember, this is for
demonstrative purposes alone. So create a realistic
image of an ancient Indian coin putting on a red dress
adorned with good ornaments. So we click on Submit, and we allow the co pilot
AI engine to understand our prompt and generate
image based on that. So it has generated our image. And as you can see, we
click on any of them, and this arrow, we
can clicking on that arrow to be
scrolling our image. Assess all of them from one particular window
from this window. Then assess all our image by clicking on that triangle icon. So we can go back to
select them if you want. If you don't want to scroll
with this triangle icon, you can always go back to
our design and select them. So that is the image
you've created. So depending on
the intended use, we select the one that best
suits us and we download. Remember, if we want
to start a fresh page, you can click on new topic. If you want to edit our prompt, you can click on that
mark to edit our prompt. Maybe to change some
of the attributes like the appearance or the action. We can always do that from
that edit icon on our prompt. We can always go
back to our previous creations by clicking on the chat section and selecting the image you
want to go back to. So remember, we want
to edit our prompt. If we want to
customize our images, we can select it and
click on Customize. So when we click on Customize, it will be giving us as to the Microsoft or it
will be taking us directly to the Microsoft
designer platform. The Microsoft designer,
if you're using Cava, you'll be familiar with what the Microsoft
design is all about. So it's an image
editing platform by Microsoft where you
can edit your images, add text, add whatever, replace background or remove background or image
editing platform. So we can select template stars, the template stars are available
from the left hand side, the top left hand side,
you see the template tab, it shows us different templates
and different functions. So that is it. Media, if you want to upload a media from your device file, you can click on my
media to upload them. Visuals if you want to
change your visuals, pick a visual, you can
click on the visual stab. Generate the visual, you can
generate a visual if you want to use your AI engine to generate a visual
if you don't have any. Then the text session is
for you to add a text to your image or, you know, adding text to images. That is
the text section. Now, the brand kit, if you have a brand van and you
want to, you know, use a unique style, you can just pick a brand kit and automatically,
everything will just change. But the font the font style, the color will change
automatically. That is for the cropping, if you want to click on
the crop icon and adjust the images as needed,
the image tools, there's a tab for the image
tools, as you can see, you want to remove
the image background, you want to erase, you want to block background,
you access it from there. On set background, pick seltOTols are available
for other function. You want to add a page, bottom, you click on AD page at the
bottom, click on AD page, then to give you or to bring out a new fresh page
for you to work on. So if you have any images, you uploaded your image section, you can bring it down here to start working on
it. So that is it. If you want to delete, you can just click on
the three dotted line by the top right and either db create the page or delete the page, just
like we've done here. And that is editing our
images from co pilot, tread to Microsoft designer
and saving it after editing. So that is how we can
use Microsoft copilot. As you can see, it
performs two function. We can generate images and
start editing our images. So just two in one
kind of function. That is what the copilot, Microsoft AI image
generation tools are about. It's a very good tool for
creating high quality image, as you can see the image we
create are high quality. And if we want to upscale
them, we can upscale them. So in our next video, we are going to be introducing another powerful AI tool
for image generation. So I stay tuned and see you in the next video.
Thanks for watching.
10. Module 8 Introduction to Playground AI: So, guys, in our browser, we enter layground.ai,
ayground.ai. And when we've done
that, we patiently wait for the interface to open. Playground.ai. So that is it. This is another platform, another powerful
platform you can use. So we click on signing in
to get started for free. If you have an account,
we click on Login. So since I have an account,
I've created an account, I'll be clicking on Login
and continue with Google because I created my
account with Google. So I click Continue with Google. So I choose the Google account I used in creating my account. And once I've done that, it
will take me straight to the AI image creation interface. As you can see, that
is the prompt section. For you to enter your prompts. Just like we've seen
in other AI platforms, that is the prom section. We go to the Explore. At the Explore tab, you are going to see
some creations by other community
members who are using this playground or AI
image generation platform. So as you can see, some of these images
are made in Canvas. So in this AI platform, you have access to
create images in Canvas. As you can see, some of
them have their promptier. Those ones that are not
made in Canvas have their prom description
attached the image where you can just copy and
edit to generate similar images or get an idea of the image
you want to create. Now my creation tab has to do with the migration tab has
to do with your images, images you've created
with the platform. These are some images I created
and you see your likes. The community members that have liked your images
are appearing there. You can select your image if it's previous images or images. You can also delete your images from any of these
functions listed there. So depending on the
action you want, you can select the
intended action button. So now, that tab
is your account, where you can click on Canvas. You can open or start
creating images or Canvas. You can do that
from that section. So the playground AI is one of the powerful AI image generation to we have that is very free. So this is the canvas
this height look like, and they've already told us that Canvas will
be going away soon. So one thing with
all these AI image, the hallways have updates, so you have to follow the pace, keep up with the pace, so you're able to get
more out of it. Now, they said Canvas will
be going away very soon. So what that means is
that playground AI will no longer be
offering canvas. Before or from September 20. So if you're using Canvas, you have to make use of it now. But I'm not much of
a canvas person, so I don't normally create
my images from Canvas. I create it directly from
the image creation tab. So as you can see, these
are all your canvas files. You can see them here and get access to each and
every one of them. Remember, you can always delete by clicking on
that three dots to delete any of the
canvas you don't want. So you go back to
the right hand side, the left hand side, you
see other functions. Remember, you can delete. You can send or share from that by clicking on that three
dot on each of the Canvas. So that is it for
the Canvas section. So we go back to your profile section and that you can see with
a slend display. It has to do with
all your creations. Then the explore
session has to do with the images
from other members. So now click on
create to proceed with creating our images. That is where we
enter our prompt. The create button
is for entering your prompt and
generating your images. So you see the
advanced settings. If you want to have
a full screen view, we can click on that and if you want to minimize,
you can also click on that. Right hand side has to
do with the styles, the aspect ratios, we have explained what
those things means. We have the styles, that's
the image dimension. The aspect ratio, there's
the image dimension. You have the stars,
there's the models. We have the refinement section. We have image to image. If you want to merge
or blend to images, you can do that by uploading
an image in this section. That is what it's all about.
Now, the number of images you want to create for
each prompt you entered, you can select either two or
four from the list there. So depending, you can
select your image output. Now, the model, we have
the stable difftion and playground Vt version. We have the stable diffusion. If you're familiar with mid
journey and blue willow, you know what stable
difftionEl is all about. Now, what the negative prompts exclude from your image
that's negative prompt. Now, SID, you want to give a
identify it to your image. You can make use of
the SID function, the control thread,
private section. If you want your
image to be private, if you don't it to apply in the public section
like the ones we saw in the Explore
section of this platform, if you don't want your image to apply in a public section, you can just set it to private. So that is it. That side
is for the filters, you'll be seeing
more filters from that three dotted circle that is like the
Olympic circles. We'll be seeing more presetrs from there when we
click on that icon. So entire prompt, you select it, and the models who appear in the images based on
the prompts you've entered. So your feed section has to do with images from
other people you follow. That is where you see images from other people you follow. So based on their preference or scale or
preference of choice. So we are going to
be seeing how we will create a image
with this platform. This platform is very flexible and most important
thing is free. There is no limit to
their free version. So you can always create
many images with it, but just that some
functions requires you to upgrade your plan from the
free plan to a subscriber. So but with the amount
of free access it has, we can create good
quality images, just like you can
see from the screen, the images are of high quality and well described
and well detailed. So we can always use this
platform to create your images. So we are going to see how we are going to do that
in the next video. So I stay tuned
and see you there.
11. Module 9 Image Generation with Playground AI: Welcome B guy. So in the playground or AI
platform, we'll be seeing, like I said, in their slot feed, we've seen some images
created by community members, and we can copy each
of their prompt to edit to get similar images
in case we come across an image we liked
and we want to use the same model or the same preset to
create our own image. We can just copy their
prompt and head over to our clicking the Create button
to our creation interface. So when we click on
the Create button, we start creating our
images by entering our prompt in the
prompt bar so as usual, we can describe our images, but we can go over to copy
the previous ones we've used. For example, we can
copy this prompts we just use in Microsoft
copilot and head over to playground.ai to use the same prompt and see how the image will turn out to be. So once you've done that,
we've entered our prompt. And if you are sure that parameters are okay,
we actually concrete. As you can see,
aspectits been set, preset has been choosing. So now, this is a realistic image created
by playground.ai. As you can see, all these AI platforms
have their own settings, have their own models and
their own appearance. Their image turn
out differently. So this is different from the one copilot created
for us initially, but it's still the same prompt. We can just use like
I said, the arrow, the triangle arrow beside the images when
we click on it to scroll and assess
each and every one of them instead of going back
to pick them individually. So when we click on
one of the buttons, we can either download
or create variations. So when we click on
Create variations, the AI platform will create different variations of the
selected image of choice. So remember, we can
always do that, or we can always delete the image by clicking
on the delete icon. So for example, let us see if we can tweak some of the
features of the images, but make sure that your
parameters are okay, your aspect ratios are okay. You selected your aspect ratio, your desired or
preferred aspect ratio. You prompt is okay. Your refinements are okay, your presets are okay, and every Latin is okay. So you make sure
you've done that, and you're uploading an image. If you're uploading
an image to merge it, you can also upload, make sure the image
has been uploaded. But you're not doing
that in this video. That will be for
subsequent video. So number of image
you want to create, let's switch it to
two instead of four. So whatever image you are
creating now from our prompt will give us a two outputs
instead of four outputs. So the custom exclude from image is like negative
prom so we can just enter. Although this one has been
set automatically for us, so we can leave it there,
and the private icon, you can start to like unlike in Leonardo AI is
for a paid plan, this is for a free plan,
so we can also leave it in the private section if you don't want it to appear
in the public session. So when we click on that tab, we can see you can
edit it in Canvas. Other functions are also
available for each of the images from that tab
action button tab we created. As you can see, there
are many. You can submit it for Daily team,
edit it in Canvas, in paint, use in image image, use image settings, create
variations and or whatnot. So that is it. For the
sake of this class, I think we'll be demonstrating
one of those functions, and that is creating
variations. So upscaling. So when we click on upscale,
like we have just did, it will generate a
high quality image of our previous image
image we are upscaling. So let's patiently wait
while it has done that. As you can see, these
are the same image, but they are of different
image resolutions. The first one has a
lower image quality unlike the last one
we just created now. And as you can see,
this is the last one we created the creative upscale we did from the previous image. So you can see this you are
seeing the before and after. So as you can see, check the I. You can see the difference. You notice the difference
from the facial descriptions, the facial features,
you will notice the difference from
the before and after. So upscaling gives us higher a more higher
solution of our image. So it's always essential we use that feature when
creating any image, especially in our projects with AI image illustration
tools, it's good upscale. So as you can see, sometimes when it gives you a known error, you want to ups the
platforms unknown arrow, all you have to do
is just to refresh. Now, like I said, the three
dotted Olympic like line is for selecting different
theaters or different styles. Just like we did in Leonardo AI, we can select a
dream shaper model, cinematic L Disney Teams model
or whatever model that is available based on their
own platform preference or their platform style. So you can always select it from that three dotted circle, sorry, three circle icon
beside our prompt bar. The three circle icon
beside our prom bar is for other theores or other
model types or model stars. So let's not forget
we can select each and every one of them
depending on our intended use. As you can see, I
changed the model style to dream shaper, and
I click Concrete. As you can see, it has given me another style of my
image, as you can see. It's a dim shaper
stylize selected, and it has given
me another star. A look of my prompts. So as usual, you can start to download or you can
decide to create a more higher resolution
by clicking on upscale or create different styles of that particular image by
clicking on create variations. Upscaling is for creating a higher quality image
of your previous image. Variations is for creating different styles of your
particular previous image. So that is how this function works in image illustration
with artificial intelligence. If you're working with AI
tools for image illustration, those functions, that
is how they work. While you click on
each of the images, you can decide to upscale or you can set to download
it if you are satisfied with the outlook, you can set to download
it immediately. So that is it. As you can see, where you go to your creations, like in my creation tab,
you will see the images appearing there as
images have created. Explore tab, you will
see other images. Like if we submitted this
image as a daily team, we are going to be
seeing them there in our community or in the community section
of the Explore tab. We'll be seeing it our
creations alongside with other people's creations. So we can always go back to our platform to create our image by clicking
on the create tab. The image creation window, we can always access it anywhere by clicking
on the create tab. So that is how we can use
this tool to generate images. So, guys, Stay tuned, we learn how to create consistent character
images using AIT. See you guys in the next
video. Thanks for watching.
12. Module 10 Consistent Character with Imagine Art Intro : So in this video, we are
going to be introducing another powerful AI
tool for creating consistent character
images without using SID. So, guys, stay tuned and see how we are going to do that
in the next video. So, guys, in your browser, you enter margin dot art. Now, when you've done
that, you patiently wait for the interface to open. And when it does, you
see an interface like this that says image
AIR generator. You can see they have the
block, the community, the Affiliates,
API and creators. So this is the
prompt bar where you can start into your
prompt if you want to. And you can also go to the website to see
other functions it has, like let's image,
AI video generator, real time generation,
and all of them. So for the proposal this video, we are going to be
signing in by clicking on the signing icon
at the top right. So when we do that,
we'll be asked to either contribute
our Google account, contribute Facebook,
contribute our discode or enter our email to continue. So for the purpose
of this video, we are going to continue
with Google account. So we can continue with Google. We have to select
our Google account. So when we've done that, when we selected the preferred Google account in case you
have more than one, now when we've done that, we patiently wait for it
to synchronize with the platform to be able to
grant access to the interface. So lick and continue
signing to imagine Art. We click on Continue
and patiently wait for the website open. So once you've
alerted ourselves, we see an interface like this, we have allowed access to
the imagine art AI platform. As you can see, there are various features
like this to image, AI video, ideation
and image remix. So we are going to be seeing
the functions of all these. So the left hand side, we click on the text Image tab to know what it's all about. So like how we created images
with Co Pilo, Leonardo, Playground, and Dali, we can just start creating our image here by entering our prompt. Selecting the style,
the number of images, the expect ratio, and
click on generates. So that is it. And the
next one is AI video. If you want to
change our images, generate videos from our images, I will be demonstrating
that subsequently in my next class.
Watch out for that. So that's the function
of that AI video tab. Then the idea want
to create ideas, craft ideas from scratch. We can do that by clicking on there product tab and we
start creating our product. The image remix, if you
want to edit our image, at some features or
tweak our images, we can upload our
image and start working with the image remix up. Now, the Canvas
have to generate, we have a project
we're working on, and we want to group them into one single project or work
in one single interface, we can canvass and
start generating our images for our project. So that is it. The
Canvas section. Now if we proceed further, we see the mini apps. Now, the mini apps
are other function integrated with image
real time generation. Like I said, as you
are typing your prom, the image will be generating
until it reaches the end. Creative upscaler is to get a high quality image like I've told you before, upscaling. Character consistency
that's the main mode of contention of this video. The character
consistency, you can see that in our sunscreen video. Thanks to email to generate
image from our prompts. So the Monday subscription tab, as you can see, they give
us 50 tokens for each day. So once you absorb
the 50 tokens, you cannot create image agent the next day except
you are a subscriber, so we can do well to
subscribe to any of the plans and have access
to unlimited creations. As you can see, they
have the standard, the premium, the professional. You have 50 token free. Remember, when you
create an account. Now, if you want to
have more information, and go to the frequently
asked question tab. Each of the plans have their functions and their
features and limitations, too. So they frequently ask
questions to give you more information on what
inquiries you want to make. Now, this is your curret
session where you can sign out of the platform. So that is simply the
imaging at AI two. You can upgrade by clicking on the upgrade tab
at the top right. The moon icon is
for you to change your font change the scenario, the screen mode of the screen, what is the moon icon there. So this is one of the
AI tools we'll be using to try the character consistency
generation with AI tools. So, guys, stay tuned and
see how we're going to do that. See in the next video.
13. Module 11 Generating a Consistent Character Image with Imagine Art: So in this video, we are
going to be creating our consistent character
images with Imaging art. So guys stay tuned and
practice along where this is going on so that you can see how we
are going to do that. S in the next video. So, guys, when we log into our
Imagine art account, the next thing we are going
to be doing is to scroll down at the bottom left
and click on Mini Apps. When we click on the
Mini App tab we see an interface like this
where we have the tools, then we go over to character consistency and select on it. When we click on it, allow
the interface to open, and we'll see an
interface like this. So this is a dashboard
where we'll be creating our consistent character
images from here. Now we are going
to since we have understood the parameters and all the terminologies involved, the next thing we are
going to be to be setting them now to
create our image. So we click on the CAL
tab at the top right, and when we click on it, we see all the stars, the
models available. Now, depending on the type
of image you want to create, that will guide
us on the type of model or preset style we select. So now we want to create a
three D image kind of art, so we are going to go over to Disney and select
that preset style. Now, we click on Disney, we've selected our preset style. That is the preset model
of image to appear. Then the number of
images to be generated, we select two since we are using a free version
that is limited, we have only 50 tokens, so we are going to
be selecting two as the number of images to be
generated for each action. Then the aspect ratio
that the image dimension, depending on our
frame rate, we select the aspect ratio we want
our image to appear. That is 16 by nine,
because we are going to be using these
images for video. So we select a 16 by nine, then the source image, that is where the constant character
technology comes in. So we are going to upload
an image whereby any image we create to carry the face
of that image we uploaded. So we're going to
click on it, go to our five folder and select the image that we want to create consistent character
images with. So when we select
and click on it, we click on Enter,
then automatically, the image will now load
in our device dashboard, as you can see, it has
uploaded the image. So now, remember to say the
face accuracy to the extreme. And when we do that
with a slider, we click on Advanced Settings, the advanced letters to do
with your negative prom. Things you don't want to see in your outcome or your image
output, product output. The attributes you
don't want to see. So I'll entering some
of them like blur, ugly shaky body parts. So it is not limited
to this negative prom. You can also well play
around the then Sid. If I want to give a unique
identifier to my image, I can enter a SID number for it, but that wouldn't be necessary
because I don't need that. As you can see,
different numbers result to a new variation of
your output image. So that is what
SID is all about. Like I said earlier when I
was introducing terminology. So once we've done
that, the next thing is to go and enter our prompt. Then whatever prompts or
command we enter will be created based on the image we uploaded as a consistent
character image. So we are going to be entering our prompts for us to create a consistent
character image with it. Now, the next thing we are
going to do is to create a pixel style image of depending on the type of
image you want to create. Remember, taking
into conclusions some models like objects, action, preset style,
and attributes. So now we are entering our prompt and we are
trying to create an image. So we are creating a
pixel style image of young one Ariana
playing in the field. Remember, this is for
demonstrative purpose, so it varies depending on the type of image
you want to create. But using this same method, you can create as many
consistent character image as you want for your project. So you'll be needing
any other tweaking or carrying a seed number or copying another SID number to be used or used in your
prompt. You can't do that. This method is the best. So
we've entered our prompt, create a phaser style image of a young girl named Ariana
playing in the field, putting on a red flowing dress. Birds are seen
flying in the sky. So like I said, you
describe your scenario. Now we click on Generate
and allow it to generate our consistent character images based on the image we upload, the source image we uploaded. As you can see it has generated two images because
the number of images we created for our
output is to be selected two as the
output results. Now, if you watch close, you will see that the facial
expression of our images, of our output image is in line, or it resembles the source
image you uploaded as a consistent character
prototype or sample. And as you can see, it has created different
variations of that image. Resembling the source
image we uploaded. So depending on the type of
project you want to create, we can as well twitch our
prompt command to suit whatever interest or
whatever style or whatever function or pupils our intended project is serving. So when we click on
the selected image, we examine it and we
can use the arrow to slide or to go through
the output generator, like I said earlier in our
image generation classes, so when you slide, you
can see all of them. Now, you can start to download by clicking on the
download button at the top right or delete by clicking on the delete
button, like I said. Now, if you want to enhance, you can upscale it. You
know what upscale means? I've explained that in our
image generation videos. You can upscale,
copy the SID number, I once you use it in your draw, but it's not necessary since you have the source image already. So this is the easiest method to create a
consistent character. So you click on the Download
button and download your image to be used in any of your projects
you are working with. That is it. This
is how you create a consistent character image as easy as without having
to copy a SID number, or like we do in mid journey, we copy a SID number or
copy the image address and attach it at the end at the beginning of the
prompt before we enter. No, you just upload
your source image. Once you've uploaded
your source image, enter your prompt
and whatever object is being presented
in your prompt will carry the
facial expression or facial resemblance of the
source image you uploaded. So now we want to
switch our prompt to create another
image using or with the same image resemblance
of our source image. So we want to create a different scenario and we
have to edit our prompt. So now we are creating a pizzel style image of a young yen Ariana in the vegetable
garden, watering the plants. He's putting on a green dress. So we've changed the action as well as the attributes,
some of the attributes, the actual we created was she's now in the vegetable garden, watering the plants, and she's
putting on a green diss. As you can see, it has generated another different scenario
of our organic objects. It has generated
another different scenario of our object. And in a very different
environments, different attributes. So we can click on it, toggle between any of the outputs and select the one we want to
use in our project and click on it download
boarding at the top right. And once you've done
that, it will save our consistent character
image generated. So that is how we generate consistent character image very easily without having to start copying our SID number or working over stressful ways of creating a
consistent character. So if you're creating a
project that has a character, to be used or to be re
emphasized on your project. This is the way you
can just create consistent images of
that same character in different
scenarios as you can. So we are going to change
I give another scenario. We want to create
another scenario. Now, we are going to be
editing our prompts, maintaining the reset model
style and the objects. Remember, I didn't
edit the pixel style. Style. I also didn't edit the
girl, that is the object. So what I'm editing
her actions and her attributes because
we want to see her perform different actions
in different scenarios. So now we are saying she's
listening to her grandma, telling her bedtime stories. So we are creating
a pixel style image of a young girl named Ariana. Now she is listening
to her grandma, telling her bedtime stories. She's lying down on the bed. She's lying down
on the bed with. So remember, you edit your prawn very well and check
for corrections. So Remember, if you are not good in describing scenarios
for your prompt, remember you can use CharPT
by going to harp.com, like I explained earlier in the image illustration
videos I did earlier. You can just enter
prompt CharP to help you generate image an image
prompt for your project. So as you can see, I've
entered the prompt, she's lying down on the
bed with the curtains open and the moon
shining inwards. So I want to edit it further. The grandma is sitting beside her telling her bedtime stories. So as you can see, I've edited my prompt to suit the type
of image I want to generate. Once I'm satisfied
with my prompt or once you are satisfied
with your prompt, you click on the
generates button and wait for your
images to be generated. So we patiently wait for
image to be generated. Remember, I selected two, and that's why it's
giving me two outputs. So the images have
been generated. As you can see, the facial expression
is being maintained. Now, if it's a multiple
character, you observe something. Here is a multiple character, and it is giving us a picture picture
the presentation of the character we uploaded, the source image we uploaded
for both characters. How do we do this? So you
patiently watch my video for consistent character
image generation in a dynamic scenario. So I'll be doing that video later, so stay tuned for that. Now, we click on it as usual. We click on the Upload button at the top right and download our image once you
are satisfied with any of the outputs that
are being generated. So once you are
satisfied with any of the output that is being
generated. Now, this is it. That is how we create the
consistent character images. We go through it, make
sure our prompt is okay. Once we have edited our
prompts, make sure it's okay. Now we can proceed to
click on generates. And once we've done that, we can patiently wait for
our image to be generated. Now we want to edit our
prompts a bit further, want to edit our art roots. So we are giving the
grammar a gray hair. Now we entered we
edited our prompts, gramma with gray hair
sitting beside her. So patiently wait and see how
the outputs will look like. Remember, this is what
demonstrative purposes. You can always edit or change your prompts
as it suits you, depending on the preference
you want and depends on the purpose of the project
you're you're working on. Now, as you can see, the
grandma now has a gray hair, and kind of they have given Ariana the men's character
a kind of a blonde hair. So these are some things that the consistent character image in a multiple character scenario will have
to take care of. So guys tating for that video, I'll be uploading it soon. So this is how you create a consistent character
image for your project. You click on Download and download whatever one
that suits you best. So when you go to
the top right you can see the scenarios of all
the images you've created, you can always
highlight on them. As you can see, this
is the first one we created, previous
one we created. This is the first
one we created. So you can always go back to them by
clicking on that button at the top right of your project or the top left of
your project window. When you click on it,
it will give you a drop down of all the projects or all the images you've
created in sequence. So you can just
click on any one of them and tweak or
change the prompts, depending on the ones
you want to edit. So that is how we create a consistent character image
very easy with imaging art. In our subsequent video, we are going to see
other methods we can use to create a
consistent character image with artificial
intelligence, too. So, guys stay tuned and see you in the next video.
Thanks for watching.
14. Image Illustration with AI Outro: So, guys, we've come to
the end of this class, and if you missed any
steps in the video, you can always go
back to check them. Now in a net class, we are
going to be discussing AI video creation in details. We're going to be
talking about how to create promo videos, dramatic videos with voice over. So, guys, stat you see in the Net class.
Thanks for watching.
Ikechukwu Prince, Digital Product Designer
