Transcripts
1. Course Trailer: Artificial
intelligence is one of the advanced
technology to create a content or images or videos. So with this course, you
will learn nado.ai platform, all the features
from this platform. Step by step guide,
you will have it. You will do your
practical experiences for your own artwork. So we will be covering all the major sections
from this platform to make you confident and master for the AI
content creation. So we'll be covering
image to image creation, prompt to image creations, even image to motion clip also. Including you can also train
your own model to make your specific content and keep a consistency
in your creations. We will be including
how to add an element, how to remove images, and real time
generation is one of the interactive features
from this platform, and there's a lot more. So each section, it's giving you advanced technique to create
a interactive content, which can take your work or your creation
to the next level. So what are you waiting for? Let's begin the course and start creating
artwork. Thank you.
2. Course Introduction & Get Started: You will come to
the nardoI website. This is the website
of this platform, where we'll be generating
AI based content is one of the interesting platform
and to explore more. So after coming to this website, you will be clicking
to the launch app, which is on the right top side. So you will click
to the launch app. It will take you to the screen where you will be able to log
in. So this is the screen. You can sign up here or
sign in with the help of like Google or
Apple or Microsoft, whichever platform
is good for you. I will be signing
in with the Google. I have been used
the e mail address, which I wanted to use and
as a sign up for free. Once I click and
select the e mail, it will bring me to this screen, which is a dashboard
of this platform. Now I can start using. I have been used this profile. Initially, you will receive around 150 coins for
new free startup, so you can which
will be refilled after every 4 hours or within a day depending
on their system. You will have some free to opens on daily basis to
create a content. So and here we can begin. So we'll be discussing the next features and the starts in the upcoming
sessions. Thank you.
3. Leonardo AI Platform & Dashboard Introduction: There are a lot of
platforms in the market. We will begin with
the very famous and the top platform,
which is nado.ai. This platform gives us so many features from
prompt to image generation, and we will walk
through that what is the prompt and how to
create image to image, and then animate to your artwork with
the very simple sap. Some very premium
features this platform has It really can give the best opportunity
that you can enhance your content from very
normal to the very advanced. It has everything which is very new when it comes
to AI generations. We will be using here for
architecture purposes. You can use it for
your designing, your modeling to your brand, and it has a lot to be explored. So we will start with this
platform, which is rep. This is how its
interfaces look like. It has so many features. So on the top, you
can see this is the dashboard after the log in. It has a real time canvas. Means at the real time, you can row and you can generate the content
with the help of AI. So you will not need to draw
some artwor some elements. AI will be generating it. Similarly, we have a real time generation on the realtime, you can see that what
AI is creating for you. Then the motion, how you can
animate the image creation. How many ways we have
to image creation, how many prom types
we have up scalar. These are the features
which this platform is providing is a top. There are some advanced
features also like to train the model and to
create the model which you want to be
consistent to your artwork. This is all we will be coming
in the coming section, we will be discussing further. At the left hand side, we
have a lot of features. These are six main
features to this platform, which is AI in
generation, motion, real time Canvas,
real time generation, Canvas editor, and
universal Apple scalar. These are the features, which will be walked to step by step. There is a lot more to explore. This is the home page
of this dashboard. After the login, we
come to this dashboard. Here you can see the community
creation which community is creating and
sharing as a public. Most of the content which is public means is being
created for free for us. This is the free content, which you can download it, which is loyalty free or free to use for
commercial purposes, also, and which has some limitations in terms of resolution and
the size is a one only. You can come to this artwork and here you can
use it for free. Step by step, we
will be discussing about all the features
of this platform. The homepage, you can see, it's for community and it has bi scale artwork
and the motions also. Ubi scale artwork, we will be discussing that some artwork, which is being enhanced. So we can say it's
a pictures images and we have motion means video. So at this platform, there are two type of content
image in Video. So we have images and the clips, which is up to 4 seconds
to this platform. So here we have some
premium features also, which is enabled because
this is a premium platform. I mean premium subscription, which I have been using
at the time being. So this is they have
some premium features. So I have apparent. Subscription at the time being and I have all these features, which we will be discussing
to this state by state. Mainly this platform is
based on to the tokens. Every month, whatever
the subscription you have or if you're
using for free, you will have per
day 150 tokens, and there is a limitations
that every generation, every use of this platform
consume some tokens. So this is very brief about
this platform and definitely, we will be walked through
the step by step. So we have discussed to the home and this is
the personal feet. Personal feets is something that the content which you
have been created, so it will be displaying here. So this is some of the
artwork I have been created. So these are the
two main features. So we will be discussing further step by
step for the other features. Okay.
4. First AI Image With Auto Prompt: Welcome back to the platform. Well, Today, in this session, we will be creating the very first AI image with all the premium
features of this platform. Now we will see that
what we can create and what is the best
available options we have. After the log in
to this platform, I will be clicking to
the image creation. The image creation is giving me this platform here that I can write a prompt here
to create a quant. This is one of the
options I have. So the second option I
have that either I can ask this to randomly create
a prompt for me, right? So I will try to ask them to create a randomly
create upon prompt. So this is the randomly
automated prompt is given by this platform. So this is all the detail
have been given to us. So now we will see what
features, what options we have. When it comes to the preset, preset is the predefined
features of this platform. I would like to have
a cinematic preset. Means the look and feel and the design pattern
should be cinematic. After selecting this feature, which is available
for free as well. Here we have the preset
which have been selected. It's depending on that.
It's a pre set style. It's linked to this
pre set style. Keep it as a cinematic itself. Then the contrast the
lighting you want. I will be keeping
it as a medium. Here we have a premium features. Which comes to this apparent
subscription to have. In the fast, it will not
be a detailed It will not be like a detailed face x patients and the detailed
artwork to this. I will be selecting a quality
as I want a quality image. Then I will be
selecting the more. I want is for dexto. For the dexto, I
will be selecting the size is a 16 by nine ratio. If you want any other size, so you can pick the sizes like if you want
the square shape, so you will be
selecting one oh one. Here you have a slider to vary depending on
the ratio you want. Whatever ratio it suits to you, you can select here to ing. I will be using the
16 by nine ratio. Well, the large option is only available for
the premium user. This is one of the
premium feature also. Here we are using the quality is one of the premium feature. Means the detail, it will be
very detailed and the size. It's neither larger. It's like already
upscaled image. Then I want number two images. I want it to create
only two images, which is a featured image. In the free, we have
only four options. You can only create a four. I'll be gating at the
time being two images. Then one more premium feature, which is a private mode. Here I have private mode that whatever content
I've been creating, it will not be published to the community even
for a free use. I want that this content
to remain for my use only. I want it to be private. Yeah. Then after we have
some advanced features. I will be selecting
a photo rail. Photo rail means it
will be some realistic. It will be a more detailing and it's a photo
realistic image. This is one more premium
feature which we will be using to this
our first AI image. Let's see how it
will be the prompt is being given by this platform, and then we are using a
quality and this like this. So You read it. I if it does not work
for you, this prom, definitely, you
can pick the prom which you would like to, right? I'll be creating
generating this prompt, which is our first AI prompt with all the premium features, which we could select
for this platform. And this premium features, even we can do
more on top of it. It's I mean, it's all the
premium features which we have. But after generating, yes, we can take it to
the next steps also. So that we will be discussing into the. It will
take some time. I will forward it. Yes. So it's being created something forest with saying something that was
auto generated, so we could stick with our style which we want or
we can let it be like this. You can see how detailing
it is. I click to the Zoom. I can see that that how
detailed artwork is it. Here you can see it is
a very night forest, and then some flash,
which is fiction. I think it's very based
on the fiction content. If you want to be rephrase it, change it, so here we
can rephrase it also. But yes, this is our first AI with all
the premium features. You can use it and you can
simply download it from here. It has an option here that you can download this. I
will download this. If you want, you can do a
lot more features are with this so that we
will be discussing into our upcoming
sessions. Thank you.
5. Image to Motion Clip: Well, in this session, we will be talking about
image to motion, that how you can generate the motion clip with
the help of the image. You can animate any image which
you want to make a video. Yes. With the AI,
we can do this. The platform which we are
using for the AI generation. Once we come to the screen, here we can see the
option is called motion. In the motion, Um, so there are two, three ways from this platform that you can make it motions, either you come to
the home page or click to the motion here, right? Or you can click from
the AI tool from the left side panel also motion or any image which you have in your community creation or in your personal feed also. Lex there is an
image which I want to animate if I scroll from
the community creation. So Lex I want to make a
motion clip for this image. For doing it, I just simply
come to the di post, and this is the option
is here image to motion. Once I click to this, then it will ask me all the
options to do this. The image which
have been selected, I can even select any
other image as well, either from my creations
or I can click CMre. Once I click to the C more, it will be showing the uploads, which I have been done
yet or my generations. Means your generations from
your profile account itself. Or either you can select
from the community field or your followers like to whom
you are following itself. So I will continue with
the community creation, the image, which I
have been selected. The same image I have selected
from the community field. This is one of the
options we have. The second option for doing
it, the motion strength. The motion strength is something that you want to know
that at what level, the image should be moving
or should have an animation. For the motion click. I will try to keep it as
much as motions as possible. Then one of one more option, it's for only premium
accounts with this platform, which is private mode. It will be a private
and the motion will be as much as
possible with this image. It will consume 25 tokens to generate this motion
as simple it is. I will click to the
generate motions. The video generation have been started and once
it will be completed, so it will be shown to
the personal field. Right. So I will right I will come to
the mi presional feeds and within a few seconds, it will be showing there. So this is one of the options. Wherever you click it will
take you to the same screen. I if I click to the motion, so then it will
be bring to me to the same procedure
to create a motion. If I click to the left side, also, it will make the same. Meanwhile, we'll be going to check I'm going to check
my personal feeds. So for now, the motion, which I have been created, it's being created already. Now let's see how
it's look like. Yes. You can see the motion have been generated
to the selected image. It's very interesting. This is our first
image to motion. Clip and Nevin, you
can simply download this from this
download button here, you can see that what
resolution you have been and how many the
versions you have. So, this is very
interesting and similarly, you can select any image you like either from
your computer or from the community and you
can create imagination. So see you in the next session.
6. Image to AI Image: We will be discussing about
very creative feature of AI is that you can create
the image to AI image. Yes, this is the magic that you can use your image as a model, and you want to imagine
and you want the AI to generate your imagination based on the image that we
can give it to AI. Well, this is the
platform leonardo.ai, and it has a very predefined
feature with this. So for doing it from
image to AI image, we simply come to this platform, and we will click to
the image generation. So for the Image generation, as we have been, experience that we were generating images with
the prompts, right? So we were writing a script or a story related to the image
and image was generating. Now, at this time, we
will use this feature here at the prompt site
with the prompt such par, there is an icon
is called image, so you will click on it. Here is the image guidance. You need to give it to the AI
that what exactly you want. Either you want to use
the style of the image, means the design pattern
and things like that, or you want to use a corrector. First, I would like
to click on view More to see that what
options we have. Here we can understand
all the features to make AI image with
the help of any image. So we can detect the colors of the image
to create a new image. This is the option if you
want to do like this. So the second option, we
have a style references. Style references, I
will keep the design itself and the coloring to
make the same design concept. That is content references. In the content references, you will keep the
content and you will enhance the picture, right? So it will use the input and then we'll generate
it based on the color. Let's say, like if you
have a furniture and you want to recolor it and imagine that how
it will look like, so that is really very useful. Well, the fourth is the
correct references, which we are about to be trying. In this corrected references, we will be giving a
sample correct to the AI, and we will let the AI imagine
what we will ask to do so. So the third, we have
a depth to image that you want it to be a
detail to the image, and then we have to image that you want to
keep the same image, and to make to make enhance
to the image itself. And then post to image
is also very interesting that you can give a person
to be a unique pose. And then you want the new model or new corrector to
pose the same way, but the corrector
will be a unique. And then we have text
to image as well, and that is very simple
that you will be writing an image on any black and white, you know, background, and you will the AI to imagine
as per your prompt. I will be selecting to
the corrector references. Once I have been selected to
the corrector references, then I need to input an image. I will be using from a community as it's
allowed to be used. I would like any
corrector which we can imagine for the batter. So I will be using this image. I have been selected the
image from the community, which is free to use, right? And then now I have
selected the model and the image which
I want to use it. So first of all, I will click
on this Mate back again. Here we have very
important feature, which is strength
strength means, at what level you want the model and the corrector to be the same in your AI image. If you want not resemblance, like you want it to be
very minor look like, then low or medium or high. I will try to keep it high
as I want exactly I mean, very nearst modest
itself, right? If I want to change the feature
like the image guidance, so I can take it here to the corrected references
and I can add it. And if I want any other image, I can click here to
the image icon and select me to select any
other image itself. So I have now I
have been selected. I have selected what is the level of the corrector
will remain the same. So now I need to select
like what reset I want. Right model will be the same
as I have been selected. So now I have an option. 30 is a portrait perfect mode, like some realistic concept. I will go for this option,
by the way, for this time. Let's say how it
would look like. So then I will keep the preset style
depending on my preset, and I will keep the
contrast as a medium. I will generate as
a quality image so that this is the
premium feature for it. And I will keep it 16 by nine ratio for like
for dektop background, and I think it's It's a good
to be has a good quality and the resolution will be so
I will have two images from it and I will keep it as a private mode that is a unique. On top of it, I will be
selecting a photo real to be have a more real image. And then now I will write what I want to
imagine with this image. So the woman is standing in frint of
apple store, right. With the the modern re red suit. In the time. I will be giving a hint.
Now I have an option. I I let it to
improve the prompt, so I will keep it like this. If I want as it is the
prompt, so we can select. I will the prompt as automatic, and I will be generating
two images I c to generate. Let's see how the
image will look like. So once it will be ready, it's like generating
at this time. So we have a lot of features. Even we after generating, we can do more with the
image to enhance it, so that we will be discussing into the
upcoming session that how we can up scale and how we can enhance the image itself. Even it ine the resolution, the sizes, or the
depth of the image. Even if we are using
already quality, so yes. We we will have a look to the image which
we selected before. Now let's look what we
have been generated. Yes. That is the model
which we selected. So the model will
remain the same, right? And we give the imagination to the person with
our unique prompt. So this model standing in
front of the apple store, wearing to the red suit, right? So I have been selected
to two images. The first one and
the second one. So you can see that
the power of AI, that they're keeping the
same model and giving an imagination and generating a very new concept of the image. It's very amazing. Here you
can download the image, and yes, that's
very interesting. So I hope you will try with the features which
have been viewed, see you in the next session.
7. Remove Background With Ai: Talking about how you can remove the background of any
image with the help of AI, which is very advanced, and we will be using for this the same platform
is leonardo.ai. When you come to this platform, the only limitation this
platform has that you can only remove the background to those images which you
have been generated. It means the images which is in your personal
feed because those are the content which
you have been generated. If you try the community, Content, let's say, this
is the community creation. If I try to click any
image from the community, and I try to remove
the background, which is only possible, you can see here that there is a three dots next to
underneath to the image. You click to the three dots and here you can see
the removed background, which is disable this option. It's not clickable because this content is belongs
to the community. So what you can do, you can download this image. If you want this image, so you will need to download this image and you
need to generate your own content with the help of the AI and then
only you can remove. So I will come to
the personal feeds, the content which have
been generated by me. So I have this content. So let's say if I
will pick this image, This In this image, you can see that it's
being generated with the help of this prompt
and it belongs to you. I will click to the three
dots underneath to the image. Now you can see that
this option is enabled because this image it's
in my personal feed, and this image is being created by my
prompt, you can say. Now this will cost two more tokens to remove the background.
I will click to this. It says that the requests have been submitted and it
will take a few minutes. I will come back to this screen back again or I will
refresh the page. So it will be removing the background to
the selected image. Now I have been come back to the dashboard
after refreshing, I will click this image again. And here you can see that it will take some time to
remove the background. As you can see that
it is in progress. So as soon as the
progress has been done, the background image
will be removed. I will click to this image. You can see that how good
quality it is because I have been created with
the highest quality. So I will refresh the
page one more time. I will refresh the page one more time to see the background
have been removed. Yes, the background
have been removed. As I can see here, right? So I have two options now, the original and
with no background. Once I click to the
new background, so the image will be loaded
and I can open the image, and here you can
see the background of the image have been removed. This is the AI and it's magic that it's
removed, it's very good. The quality of the image, you can say you can see that the quality of the
image of the background, it's good. Now it's good. So yeah, this is very simple steps and it do
not cost too much tokens. If you even have a free account, so you can use this
platform to moving images from your content. So yeah, we'll be discussing more features in the next
sessions. Thank you.
8. Canvas Editor: We can do with the Canvas, including enhancing
your existing content or doing some artwork. All we can do with
this platform here, Leonardo A I will click to
the left side Kanvas editor. It will bring the Canvas with all the features
of this platform. What will do, I will simply
click to the screen here. And I will introduce you
all the features it has. Here we can see
the select option for selecting the elements, and then we have a draw mask. So with the masking, what we
can do, we will have a look. Then raise option, we have a sketch text and
uploading an image. This is the fn side, and then we have a prompt definitely to with the
help of the prompt, is one of the amazing
tool to do things. Um, and then at the
right hand side, we have the nado that
fine tune model. Til now we are using
the predefined modules. In the upcoming, we will be
using our own model as well, and then Canvas mode. It's input and out pain
mode in pain and out paint. Which you can change it and depending which you want
to use in an output, you are interacting with the, not only the image, your input will be there
with the image as well. Then we have that what is the stre level is only one for now and how many images have every time is
supposed to be created. We use a prompt, How many images supposed
to be generate? I'll keep it for
four options now, and then the image dimension. For now is 512 by 512. Here this is the frame
actual frame size here. You can resize the
frame here as well. If you want to keep the
dimension one by one, you can keep it or if you
want any other dimensions. Let's say if you are
creating for desktop, so you will be picking a 16 by nine and similarly for others. The rendered density, you can decrease or increase the
render density of the image. I keep it uh, as it is for now as a default, and then we'll see
the guidance style. Guidance style where
whenever we are writing a prompt or giving artwork hint. What level should we follow? I'll keeping it as it is
17 for the time being. Now let's begin the
magic of this platform. I will be picking an artwork from the community
for the time being. I will come here to the
community and I will try to pick an artwork here. So I will pick this very glamorous castle
gold by mad and all. I zoom in and we'll have a look what exactly
the artwork is. At this point, we
have the image, which is unable to
fit to the frame. And it's vertical. So I will try to keep make
this image as a horizontal. Definitely, when we are
generating with the AI, so we need to imagine for that what we
want to do with this image. So I for making it, first, I will need to increase
the frame size, I have been increase
the frame size and I will bring the image
into the frame. Then after I will try to
resize it into the frame. Still the image is not
filled as is supposed to be. For doing it, I will give
a prompt a structure. And then it will be adding a structure as per
the similar image. It's a very simple pro
hints if we do it, so it will do the
magic with the AI. Yes. I got a four option as we defined that every time
we'll have a four hint. Which hint is for
you, you can pick. I will pick at this time. Like this. I think it's a good, so I will have some space
to do more at work. Now we have enhanced this image, adding a more content
of this image with this vs. What
is next we can do? One more very advanced
feature of this platform, which is adding an element. Adding an element with the
help of this mask tool, I would like to have
some birds around. I will need to select
the area which I want. Whatever area you will
select with this mask, that area will be used to add your next element with
the help of the pond. I want flying birds and I will generate
the flying birds. It will be generating
birds to the area, which I have been selected
with the help of the mask. As soon as it's being generated, it will give some options. I got four options again, which will be very
nears to the realistic. I will go for this. Two birds are flying and it's
very interesting. I will say except. This is as simple as we
could do with this image, we pick something and
we make something.
9. Realtime Canvas: Interactive feature from
AI that in the rail time, you can have a rail
AI generative, then you can you can do the
your artistic work with this. For this feature, we
have a realtime canvas. So to access the
real time Canvas, either you can click to the first option from this platform. Right or from the left side. So you simply come here and then you click
to the real time Canvas. It will bring you
to this screen we, you can start your
creative art world. I will bring to my screen down. So you have a lot
of options here. When it comes, you
can start with the image itself that you
want to start drawing, and then you can
interact to the image. Or you can write a prompt
and within the prom, you can start working on it. What I will start with, I will introduce you to the
interface first here. Here we have an option to select the image from the left side, so we can even select from
the community creations or you have your own generative. Then we have like
a Canvas option. So we can select a limited
area from the Kanvas and you can draw or you can
give a hint to that area, so that we will be
looking forward as well. And then we have a eraser, we have a selection tool
as well to either to select an element or something. And then we have a brush here. Brush we'll be discussing
is something give a very prominent hint to AI that what
supposed to be done. I will start with very
interactive option is that I want to a joint rebote phase, right? I want to start
with this option. What I want is now I want to
interact the real time s. I want to interact to
this robotic phase, and I want to draw something to give a hint what
I want exactly. So I will select the
colors, which I want. So what colors I
want for this bot, I will select this first, then I will draw how it's
supposed to be look like. So I will I will give a hint. I give a hint. Don't worry
whatever it's making. We will make it. And then I
want red eyes not very big. I want red eyes here. And I want his nose. Then I want a light to be on his on his address itself. Whatever we have been draw at. It's a hint to the AI that
how it's supposed to be. You might see that it
might not look like sharp. So we can simply click
to the instant refined, which is at the
right hand side top. It will refine, it will be
or quality perspective. I click to the instant refine, so it have been refined
the image itself also. This is one of the options. Another option to
refine it more. We can click to this
next image upscaling, it's a featured
image option itself. But I will click to this up scaler and you can see
here refiner strength. If we can do more refine and it will be consumed
some specific tokens. I'll be going to the higher
refine and I will upscale. This report, what I
have been generated. So this have been refined now
for more further options, if you want to refine more. So even you can give a more
more detailed hint like with the red lights
in his dress. I give a more
detailing for this. On top of it, let's say, if you do not like a
red light on his eyes, so simply you can
select the arrays. And you can remove
that those eyes. It will not be to
the eyes itself on. For doing it more, there is very good options here that you can select before
the prompt area. We click here and this is the guidance to
this specific tw. Whatever have been drawn, it will be followed to whatever
had been drawn to the tw. You want to fix the hint, but whatever have
been generated, you want to Another
version of this artwork. Keep the hints. I
will just click to this use the fixed set. Yes. I'll keep clicking and keep generating more
variance for this reward, and the quality is
keeping it high itself. For the more options, we have
been seen here to the left and side that you can
have variance to your. The guidance you are giving
you want it to follow as what level you want it
to follow and quality of the image itself.
Here's the prompt. Whatever the prom you will give, the word will be
reacting to that same. And the creativity strength, the depth of the artwork will and here is
the preside styles. You can select whatever
preside style you like. Let's say if I click
the cinematic, so it will be a cinematic
scene of the revote. If I click here the concept art, it will be look like
a concept art scene. These are the preset. You can even define
your preset also that we'll be discussing
into upcoming sessions. I will be keeping a
photograph at the time being, and then we have an option here that there are two options. This option is an
interactive mode. Whenever you draw and after
you complete your action, then it will react on it. It means you are in a progress, so you do not want
or a real time. I was keeping it as a real time. At a real time is interactive. That is the concept
of real time canvas. You will know that what
you are generating. Here you are writing
your prompt, like we did in image to image. In the image to image, you just have a limitations that you can only
write the prompt. Here you can see what your
prompt is generating. On top of it, you are interacting with the
help of this Canvas, and you are very interactive and controlling the features
and the options. I hope that you
will be using this and generating some
creative stuff. If you have any questions, so you can communicate and we will be walk through the more options in the
upcoming sessions. Thank you.
10. Realtime Generation: I would say it's me
generation next level. Where your prompt is not something like you
are waiting and after after creating
your prompt, then you will know that what exactly you wrote it
and your imagination. Here we have all the
real time features and the control to
your generation. We are about to explore the
a real time generation. For doing it, either you
can access on top of the screen real
time generation or from the left side,
real time generations. I'll be clicking to the
real time generations. I have been click the
real time convas. You make sure that you click to the real time generations, the second option. Here
I've been created. I have a canvas here where I can see that what
I've been writing, and this is the
place where you will be writing a prom bar, so you will be giving a hint
what you want, exactly. The prompt I want is a rule right is in the London, a young couple is standing beside this right
it's like this. A young couple is
designed, right? So One more thing is that you are not limited to your
prom to just write here. You can even enhance,
you can use a chat GPT, but here you can see that whatever you are
writing step by step, it will be, you know, you will see that how will R c. Right R car. Now you will understand your storytelling is very important the way
you write. Yes. Now what I want is actually, I want to be upgraded
to the next service. I can even go to the GPT, and I can write the script here, which I have is and I can ask
the to improve my prompt. Here, I give a hint or line
to improve the prompt, which the line I have. Then I will bring it here. Here you can see that
it has been enhanced. If you don't like,
you can come to the problem which you
wrote it depending on it. On top of it, you
have more options that you can randomly
create more seeds, or you can change the preset. I have here the list of preset. In the upcoming
session, we will be discussing how you can
create your own presets, all the models, which
you want to be continue. I will stick to the
one which I cinematic, whatever you like, you
will stick with that. Then here you have two mots. The one mod is retaining the quality of the
original image, and I will scale. In the creative, it will do some more artwork into that when it's doing a
scale. I will upscale it. I will up scale it It will
cost eighth tokens more, and it will be after
this upscaling, it will be to the personal
feed where I will have it. These are all options. To the real time, it's
very interactive. You can use even after
generating you if you want to see the
more variants to this. You can simply click
to the random seat. It will be creating the more
variants to your prompt, which you have been selected, the final you have. Here you can see the elements. I will be discussing a
bit more about this. If you want some
kids illustration like some childish
look to be added, so you can increase the
bar and it will be adding these element more If
you do not want you want some young or modern young, so you will be keeping
it to the minus itself. And the tone and anime. You will be keep increasing, so it will be towards
the artwork site. I will be keeping as it is. A fork fork art. Also you can variant to
this and the coloring book. I prefer depends your taste
the way you want it to
11. Prompt Types: We'll be discussing about how many ways we have
a prompt generation. So prompt is actually
when it comes to the generative AI or create a content with the help of AI. So prompt is something
is called a script, or you are telling a story to the AI that what needed
to be generated. So where we can use this prompt. So when we come
to the dashboard. So as I mentioned, that we have here
for this platform, we have six options, so we can generate
a content with the help of these six options, we have some advanced
option also. So two advanced options we have, and six we have. So we'll start with
the image creation. Image creation is either we can create the image with
the help of image to image or we can
create as a prompt. So this is the where
we write a prompt. Pm is like a story that what I wanted this AI to
generate for us. So let's say I say if I
say a woman is standing in the green garden
with the lake around. Right? So this is my script. So now what I want is that
with the cloudy weather. So whatever the phrase
you're writing, it might not be very
detailed or descriptive. So it's not generating. So it's not correct.
Means it's not completing your image concept. So there are so
much detail needed. So what I can use it. I can use to improve the prompt. So it will automatically
will improve the prompt, which I give it to them, right? So this is one option. The second option that if I
do not want to use my prompt and I want to generate something
a random so this option. This is the second option
to generate a new prompt, which will be randomly. If I collect to this, So
this is the second option. Now let's talk about
the third option. That you want to give the add it quickly
change your prom. So whatever prom
is being written, so you can change
into this like, you might not need to change the whole history and
everywhere into this prompt. You can change the
weather to night. Something, right? So it
will rephrase and it will bring the concept you
want in this prompt, it will be added into this. So this is the third
option, right? And describe with the AIs. So this is one of the
interesting option is that you upload a picture and AI will
generate a prompt for you, depending the AI will read the image and we'll write them. So now we have discussed
you have four options. So either you can use a randomly prompt if you
want something new content, And the second option is that Linaro AI will
improve your prompt. And the third is
you can whatever the prom you created and you want to
rephrase or re imagine, so you can use this or
you can use an image. Let's say if I use an image. So I will be using an image. To generate a content. So I will come to my
generations, right? And I will be
selecting this I to understand that either it
will write a prompt for me, depending on the image, which I have just
uploaded to yes. So I read the image
and it write there. Prompt for me. So there are four options at this
platform, right. So now let's talk about what
we can do more than this. Like, here we have to
improve a prompt, right? But what I want is, I will come to the Church
because I feel like this will be a more advanced
than what I want is. So what I want is a
woman is standing in the garden with the lake around right in
the morning view. Right. So then after,
I will tell to the Chat GPD to create a
more prompt for me, right? So prompt options. Once I write the prompt options, so it will be start generating
the prompt options. So here's a few prompts for depending on the
line which I gator. So it's a hint. It's a hint for the prompt. So now it's keep
generating for me. So now I have so many options
with the help of h GPT, and I can pick the best
one. I can read this. So it's like as Don't
breaks a woman's trend, so I will use this this prompt. Right? If I come here, I can generate a content with
this help of this prompt, which is generated from AI. So I've copied the
prompt and I click the generate the prompt which have been generated
from the Chat GPT. So Chat GPT give me this interesting prompt and
with the help of this prompt, I have been created
this creative tent. So you can see that
how detailing it is. And this is the imagination, which I have imagined
that that's what I want. So the girl is standing
latest around, and that's what it depends
that how detailing you want. If your story is in more detail, so yes, there will be
a more detail sorry. So these are the ways
to have a prompt, so that we have discussed here, and we will be using these strategies to create our content into the upcoming
sessions. Thank you.
12. Negative Prompt: Welcome back. In this session, we are going to talk
about negative prompt, what exactly is
negative prompts. So generally, what happens
that you write a prompt and it's not descriptive sometime or if is in detail or
very descriptive. So what happens is that some of the elements or
the context of the image, it comes, which you
are not looking for. So we have a feature in
Leonardo AA itself that you can write a prompt as
detail as much as it is. And the things which you
do not want in the um, in the creations, right? Either it's being mentioned in the prompt or
not because you can if you start saying that this
should not be like this, the creation so shouldn't be like this because
it's automatic, and it's it comes very random
creations some of the time. So So for doing it, so we will be writing
some key words which we want the AI
do not create it. So it could be any thing like you do not
want color in this way. The image should not
be blurry or the low quality or duplication of any ment or something like that. So for doing it experience, I will begin let's
come to the screen. So once you come to
the aro AI, dashboard. So simply what you need to do, you will click to
the mage generation. Once you cli to the
mage generation, it will bring you to the screen. So I have already
written a prompt, which is a futuristic city that so many
buildings are around. So for doing it, I want it to be a
negative prom cons. For doing a negative prom, the option is in the
advanced features on the left side down, and here you have a tog
fat clip to this and it will bring down the
negative promotion. This is the place where you will say something
which you do not want. You do not want a blur
cloud or a blur screen, or low quality, you can
say a dark dark buildings, you do not want.
Let's say, right? And you want something which you think that
you really do not need. You can keep writing
the keywords. I would say try to keep it
like a keyword concept, not a phrases, but it's
up to you the way you want as detail as much possible. So it will be the
concept which you want. So this is the place where
you making your pro strong. So the more strong your prompt
is your creation will be. Same. So for doing it, I have been written already. I and a duplication of
buildings and things like that. So as much as I think which
I do not want and I do not want to be write in a prompt itself because prom
is being written. And some of the time it happens that you generate a prompt from a GPT where it's being already written
and it's very detailed. This is the quickest way to make your prompt strong and your
creations with the negative. So simply I have been written
a negative prompt words. I will select my features or the setting which I
want for this prompt. I'm keeping the quality image, the generation
mote, not the fast, and keeping the
portrait cinematic. If you want, you can try with
different options as well. So we have a lot of presets, which press it's quick
for you. You can pick. And then I'm keeping the resolution or the
size of the image. Is the decktop,
16 by nine ratio. The size is large. And keeping the four images
for the creation, keeping the private mode and the photo real
is also selective, so the images will be nearest
to the realistic concept. This is all with the
prompt and negative. I will click to de generate. Now is to start generating. If you have seen the previous lecture that I used this prompt to generate
a four more images, and you could see the
images are down here, and the prompt I was
keeping the same. So this time, the prompt
is exactly the same. The only thing which we have
been added is negative. But definitely it increase the concept and the imagination of the
image which you want. So you keep practicing
that I mean, when you keep generating, you will understand the power
of each feature it has. So most of the time, there
are some predefined keywords. You can write. This
is a tip for you. You can say you can write a preset of negative
words like blurry, low quality or, you know, some some keywords, which is very genuine or common keywords. So the image have
been generated. So I got four images keeping the same prong,
which I used earlier. And you can see the
buildings are not dark. It's not a night mode. And it's not a low quality. It's a more high quality. Every image, it is not blurry and it's not a blur
cloud as well. So this is the power of
you can say the prompt, the strongest prom,
is the best way to take your prompt
to the next level. So I hope you understand
this and you will try this. In the next lecture, we will be talking about more features. So we'll see you in the
next lecture. Thank you.
13. Create Your Model: Welcome back. In this session, we are going to talk about
how to train your own model. Relatively, it's very simple, and I will tell you in
the step by step by. So when we come to this
dashboard is you will navigate from the advance at the left and side
panel and you will click to the train
and databases. So this is where we begin
the creating your own model. So you will click to
create your own model. I have been created
these three models at the time being, right? So here you can see that I have I have been used
these four images for this model, and
we used to one. For creating your
own, you simply need to click to
the new data set. It's a data set which you can keep a concept,
which you want. I want a let's say cinematric. Right? So I want this concert
as a cinematic scene, like, filmy and all
something like said. Mating photography concepts. So what I'm doing is, I'm creating my own
cinematic dataset, and I'm trying to train the
model as much as I can. So with this platform, you have want to
click your dataset, so you need to put the images. So you can put as
as images you want. I will be adding the image I will not upload
from my computer, by the way. So I will just simply
select the images from my own creatives. I've been selected a single
image at the time being and I will be adding
into the Dataset. After selecting an image, you can select as much as you want and then click
to the train model. I will click to the train model and then it'll asking
me more details. I will just write A here. The model type will
remain the same because this is the only model type
we have at the time being, and then you will
write the detail about the model which you being creating within this
under data set. I'll be using a resolution 768 and diffusion style diffusion
version, I will be using 2.1. Then you will need to
select a category, simply select the category which is more relevant
to your content, I will be selecting
a photography. And here you need to write
a keyword or a word, which whenever you
use in your prompt, so this model will be called it. I'll be writing the the keyword which I want to use ABGsel. So with my premium
subscription with Apparents. I have a limitation to
created ten models. I have been c already
three. I have a seven more. So I'll be click to
the start training, so c my fourth model for and then now it's started
progressing at the background. And as soon as being done,
they will e mail to us. So as mentioned here, it takes sometimes
several minutes or hours. And I will click to
the view job status. I can see the job status that what is the
status of the model, which is being just created. It is in progress. So meanwhile, we can discuss what are the
ways to use the models. So at the left hand side, we have a section
called Advance. In the advance, you can
say we have an option here fine tune models. So fine tune model
is this section where already fine
tune models are here, all the fine tune models. So we have here related
to platform models means, which this platform is
providing us to use. So whenever you are
using your prompt, you can use this specific
design concept as a model, and you can continue
your creative, so it's not limited to any. It's element, it's a background, it's a pattern design concept. So this is all we have. Then we have a community models. Community models is
for the public use, is loyalty free and is free to use for commercial
as is the public here. And you can see that what community is
being created to have more inspirations in your model and how to train and keywords
which is being used. One more thing, the
community creations will be limited to their
keywords, by the way. So what keywords
they are creating, it will be remained to the b. So your models will
be displayed here. So here, I can see that the model which I have been
discussed is being created. So this is the instant prom. Maybe, I will just simply click
generate with this model. As soon as I will click, it will take me to the screen, and here I can start creating
the concept which I want. So what I want is here. I will start writing my pro and I will be
using the keyword which I want is is a simple adject. I want to be this
very simple keyword. It's a, whatever you want. And then after you need to
select the fine to model here. So it's already being selected. If it's not selected
or if you want to select a Unique
or any other model. What you need to do SPL, these are the listed available, which click here,
select other model. And then it will be letting you to select from
your model section. So I will come to your model
section, click to view. And this is the option.
I just being did. We simply click here generate with this mode it
will be selected. Then you can select
the alchemy style, and I will be selecting. I will be keeping it
as a photography. Right? And then after I have been selected all
the necessary features. From now to left in side, we need to discuss
very general things like how many mates you want. And then, do you want
it to be public? It's up to you if
you want public, if you want transparency, if you want for photo rail, you can select these options
and the collections, we'll discussing into
upcoming sections. Here you can select what is
the image size you want. I will be selecting 16 by nine, which is a desktop and you can increase image size
as much as you want. So I'm ready with
all my details, the information which I
need with my creations, and simply I will
collect to the generate. So at this time, it will be
keeping my reference model in the image generation and the features and the options
which I have been selected. So it's two images, 16 by nine, dext of size, and the alco design style will
remain as a photography. As soon as it is ready, we will be having a look. You can create as much as
model with a defined data set, or you can add more images
to having the more detail. As much as similar images
concept you have for the model, it will be easy for you
to use in your creations, as it will be a consistent
see in your creations. So I will come to
the personal feet. My generation will be ready, and it will be into my personal
feets as is not a public. So yes. So the concept is
being created here, which I have been
generated. So yes. This is the concept, which is the model
which I've been created with all the details
and the phase cation system. I can enhance it,
I can create more. I can do it more with this. And whenever I will be creating, the detail will remain the same. So this is all
with this session, and we'll be discussing further into upcoming
sessions. Thank you.
14. Add Element On Images: Well, in this session, we are going to
talk about how to add an elements in
your creations. It's one of the
advanced feature of Denardo AI or AI itself. So when we come to
this dashboard, for adding it, we need to
navigate to the Kanvas editor. So Kanvas editor have its
own detailed lecture, so make sure that
you have watched it, so it will help you to
understand it easily. So We need to come
to the vase editor. I will repeat some
necessary things here. So the Canvas editor has
led you to addit your already existed
existing creations and do it more with this. Not only adding an element, there is a lot of features along a side to
adding an element. So for starting it, at the left hand side, we have very necessary
features for tools. At the right hand side, we have all the
functions and properties related to the elements. So I will simply
select an image, so I can select an image
from the community itself for adding an element. So I will simply select this
picture to add an element. What I'm going to do is first
unit to be ready and plant. So I'm going to add but in
the sky, which are flying. So I need to so you
can see is a blue box, which is a canvas. So either I need to
resize the image or increase the size of the mict. So I will increase the
size of the frame itself. So I have been increase
the size of the frame and the size of this. As much as I can, then I need to adjust and I need
to make sure that cloud area spatially
is under this frame, the area which I want it
to be worked with, right? So I have been bring it back, and then I can click
this lock button. So simply now, image
will not be moved. I mean, the area is being selected mean
the frame size itself. So now I can start the creative. What I want is. I can simply
click to the draw mask. So after doing this all the settings you want to the frame and image
you are ready with. So you will simply select the
door mask and then select the area which you want and where you want
it to be, add an element. So I can increase the
size of the prish, to make it more quicker and
select the mask area as my mask mask area is most
of the area of the image. I will select all the area which I want it to
be covered with. So yes, I have been
selected this all area. I want. Okay. Then simply you need to write the
keodes of the element, which you wanted to add, so
I'm going to add the birds. So birds are flying,
I will simply write and I will
click to generate. So it will be start adding
the elements on top of this creatives and only the area which I have been selected. So yes. This is how simple it is, and you can see
how beautiful and creative is being added
on top of our creatives. So it's keeping the
background as similar. As is possible. And
depending which you like, I will go for this option
and we'll accept this. So now you can see the image is the image is being ready with
with the added elements. So this is the power of AI. You can do a lot more, not only to the skis, you can add elements, and you can do a lot
more with Kanvas editor. So this is all
with this session, I hope you will be doing
some more creative with it. Thank you. See you
in the next session.
15. Upgrade Image To Next Level By Alchemy Upscaler: Well, in this
session, we will be discussing about the
alchemy up scale. So this is one more way
to enhance the picture, which you have already
being generated, and you want to take it the quality of the image
to the next level. So Alchemy is a virgin two
of the quality enhancement. So I will come back
to the personal feet to see that the content
which I have been generated, and I can pick it from my
generated content also, or I can take any image and
I can generate the new one. Right. So I will be generating the new image to let you to
understand a bit better. So I will come to the
image generation. And here I will be writing. I will be writing the prompt, which I want. Okay. I have been writing the detail, and I'm keeping the prompt bit similar to the Scottish man. And the scene will be modern. That was the 1,800 scene. So I'm keving it a modern scene to understand the difference. I'm giving it the similar
details in this image as well, the quality of the image is
a destop 16 by nine ratio, and two images with
the private mode. And the options are photoreal it's already
being selected, and my prompt is being written. The only the only thing
which I have been added is modern scene like some new
concept of this image. So this image is
being generated. The Alchemy is something that you want to be use
limited tokens as well. Because universal scalar in terms of token is bit expensive, because it upscale also, like you make your
image enlarge. But the alchemy scalar
is something you want to just increase the content. So this is the images
already being generated. So you can see, is
a very closer fee, similar Scottish man concept. Well, so here you can see
the phase expression and everything all the details
related to this image. And now you have the
option which we wanted to be access is Cam scale. You can also access from the image itself at the
right and right bottom. This icon. I led you to. So we can see here
that this option is used to refine and
scale your image. Scale means quality
of the image. Have clicked to this.
Now it's asking me that there's two options.
Refiner strength. What level I want it to be enhance the image
or refine the image? I'll keep it high. The smooth molt I made it on. These two options. Now we can see the tokens which I've been using to doing the coming
scalar is only eight tokens. But in the universal scalar, I spend around 110. I almost 10%. Of the tokens which
we are using for it. Right? So the image have
been is in progress. So I can access it from
my personal feet itself or I can use it from
him here itself. So I will come to my personal
feet to access the image and there's the difference that you can differentiate here. D are all your images and the images which
you have upscaled. And your motion fortunately. I'll click to the image
which I have been upscaled. I so here I have scale. So you can see the image
details have been increased. The face presion is a
bit smoother, right? And blurry and more focused. So now, what's the difference? So once I click the
down top down menu I can see I have two
versions of this image. The first version
which we have been generated as a sample, like a first image. And the second is
the scale, also. So even the sizes are
exactly the same. The size has not increased or the quality have
been increased. So this is the original image which I have been generated. Now I will click today. Smooth and high quality. So that is the difference. Once you have been generated, phase expiations
are not in detail. Most of the situation, if your prompt is strong, your phase expresion,
and your detailing of the image itself will
be a good and better. As prompt is directly
interact into the image. After generating, scaling is
one of the options we have. So alchemy scalar
is the best option to enhance your image
with the minimal tokens. So this is all for this session, we'll be discussing more
features in the anomal sessions. Thank you. T.
16. Image Guidance - Advance Feature: Welcome back. In this session, we are going to talk about the one more advanced
option of image creation, which is image guidance. So as we know that we can write a prompt to generate a image, and we have already know the negative prompt and
other options of the prompt. I want to make sure that you have seen
those lectures as well. So this is one of the feature that you want to
take your creations. To be more emphasize of
something your hints. So keywords or the
prompt is one of the way to tell to the AI what exactly you want
it to be generated. So at this time, I want to use the image guidance features. So what is exactly? Image guidance once you
come to the dashboard, so simply you need to click to the image creation
from the dashboard. So once you click, it will bring you to the image creation page. Here we have a prompt bar where we write a prompt exactly. We will be writing
a prompt as well, but we need to give a
guidance to the image. Means we will give a image as a guidance and we
want AI to follow it. At what perspective, so
we have different types. For doing it, simply, you
need to click to the icon, which is before prompt. We click to this icon. It's only available this feature for the limited
presets, by the way. If you will try to
use this for Phenix, it's not available for Phenix. If you select the Phonix preset, at the time, it's not
available for Phenix. So you need to select
any other presets. I'm selecting to the
portrait perfect preset is a design pattern which you
want to use for ucations. So now I can select
an image guidance. Once I collect to this icon
image before the form par, here is giving me the
image guidance options. Tree of them are showing here
at the time being and I can colect to the view more to see what other features
options we have. So at the time, we have
eight preset options. The first one is image to image, and it's for
specifically where you have a black and white picture and you want to
make it a colorful. So this is the option
which we can use for this. And the second option which we have is a style references, means the color theme, you want to continue,
so you can use it. The third one is the
content reference. So the content reference
is something where you want to enhance the
colors of the picture. Your colors are dim and it's
not as colorful as you want, so you can use this contest. You can upload a picture, which is with the dim
color and it can make it your prompt to the colorful. Similarly, we have
a caracter offences where it will follow
your caracter, the caracter you will upload. Your next prompt will
follow your caracter. A model itself, you can see. And the second, we have a
depth to image where if we can put the x ray image
kind of to make a colorful. The third is to, I will keep the image
exactly and we'll take to the next level in
depth, depending on your image. And the post two
image, post image is also very interesting
with someone, you are giving a person
which is posing for some style and you want your next creation to follow
the similar kind of course, so you can use this future. The last one, but not the least is that you can write text and whatever the
pattern you keep in your text and your prompt
will follow the text create, and I would say it's
very good for wall based or pain paint things or
logos and things like that. It's very creative
stuff as well. So I will be using a
correct reference. As I would say, is a quakes, a model training concept, which AI itself has, I'm selecting a
corrector reference. Then I will use
selecting a cocor from the community itself. So I'm selecting this cater to follow AI for the next image. So Okay. So depending, you can keep your prop
as simple as you want. And let's see that at what level it's
keeping the concept. So I so very simple and I'm keeping the improve pro feature. So it will improve my
prompt automatically. So I have been selected even if you want you
can use a GPT to you. So as simple it is, I'm keeping the options
from the N side. 16 by nine ratio for the necktp. The large image, four images to generate with
the private mode, so it will remain to the private and for the
advanced feature, photo rail is also selected for the highest quality
of the picture. And the other options
I'm keeping a same as it is for the
time being selected, what will be the creation when I'm giving an image
guidance reference as a corrector reference. So my creation AI is supposed to keep this person as a reference, a corrector, and creating
a new image for us. So there are four
images it's consuming. Generally, it takes
around 40 tokens when you are creating
directly with the prompt. But at this time it's taking 42 because there is one
more additional feature, which is we are giving
image as a reference. So as you should know that
every feature cost or consume some amount of
tokens to generate. If you're using a free account, you will have 150 tokens, and if you're using a premium subscription apparents,
like I'm using. So you will have 8,500 tokens and every time you
use any premium features, so you will it will consume
some tokens from it. So creations have been created, and you can see how
exactly and nest is the creation and the
quality of the image. Scatter is keeping the same. And that is the feature of it. So at this time, I
didn't even select a train model or any dataset. I just wrote a prompt and give an image as a hint and it's generated AI generated image with the reference of the image. So this is one of the
advanced features. I hope you will try this and we will talk about more features in the next
session. So thank you.
17. Image Elements Advance Feature: Welcome back. In this session, we are going to talk about
one more advanced feature of image generation, which is add an element. Element is something that
it's similar concept of giving an image
guidance that you are giving the hint
to your prompt, that what type of element or pattern is
supposed to be used. For doing it, let's
come to the screen. Once you come to the dashboard, you will simply take
he image generation. So that it will bring
you to the image screen. And here you can see
that you have a prom. Before the prompt, there
is the icon called image. It's not allowed to
select at the moment as we have been selected
Leonardo Pionex. So nado Pionex is the name at the time being and it
do not allow this feature. So we need to select
any other preset. So I'm selecting a portrait
preset portray perfect. So I've been selected now this
often I have been enabled. So for doing it, what
I'm trying to do is at this time is that I'm I
want to give a prompt. From will be very
detailed, right? I want some specific
design creation or a concept to the
image to be followed. For doing it, I will
click to this image and here you can see
below image guidance. There is an option for elements. I'll click to the B more. Here I can see that I
have lot of elements. It's a design is a kind of guidance or an element to give to the creations on
top of your prompt. So I will be selecting game UI. So at the game UI, you can give this element
to your creation. So at this time,
what do I will keep this element of your design
or your creation as it is. So I will just say Matrix hero is standing next fo menu, right? I'll just simply give
very brief line. A Matrix hero is the
standing netual food menu, and I will improve my
prompt by clicking improve prompt within
this platform. It will improve my prompt, and I'm giving an
element to improve it. So for doing it at
the left hand side, I will keep my features like the generation
mode as a quality. Not of fast, and I will be selecting a
resolution as a 16.9, keeping as a large image, selecting the four images, and the private more to keep
the content and photoal, and remain features
will remain as it is. So I will simply click
to the generate image. At this time, I have
given my prompt, one more additional feature which we're using
at this time is. Giving element to my prompt to follow this design
concept, right? So at this time, it will
be look like a gaming, which I have not been
mentioned into my prompt. So my problem is as simple a is. It's a story, you can say. So as simple it is, and then you can sl as
whatever preset you want. I would say this feature to giving a image
guidance or an element, it's a kind of dataset
or a model training. In your platform here. So these two features are
something very quickly, you can generate your
dataset or a image concept. So the image have
been generated. I will just click to
this simple it is. So here you can see it's keeping
the same design concept, which I have given to this. It looked like a gaming concept. It's not near to the cation. And the food menu
is around. So yeah. So this is very interesting. Okay. So this is all I hope
you will try this feature. So see you in the next
lecture. Thank you.
18. Power of AI | Image To Prompt: Welcome back. In this session, we are going to talk
about the steps to generate a prompt from a image. So it can be your image which you captured
from your camera or it's the image which is being generated
from the AI itself. So there are many ways
to create a prompt. So this is one of
the main feature of nado AI itself that you
can write a prompt. Means the system
will write a prompt for you once you
upload an image. So it can read an image
either you have clicked from your camera or it's being
generated from AI itself. So for doing it so, let's come to this
grain once we come to the out AI to the
dashboard of this website. So Simply, you can click
on the image generation, which is the standard
way to generate a image. Either is from the top menu, or you can click from the
left side image generation. So once you click to
the mage generation, it will bring you to
the standard screen where you can
generate this image. So either you type a prompt, but in this, we are going to write a prompt
automatically from an image. So for doing this, At
the right hand side, here you can see here
we have a three dots. So click to this stars icon, and here we have four options
to generate a prompt. So the first option
is new random prompt, which is very randomly, it will generate a prompt for
you. This is not an option. The second was, whatever you have written
in the prompt bar, it will improve that prop. And the third is
that you want to whatever you have been
written into the prom bar, and it will rephrase for
you basis of the keyboards, which you will just
give it to them. Let's say your prompt is talking about a morning and
you want it to be night. So so many options or so many places the
morning would have been used and you want to make it evening or night
or something. So this is the simplest
way to use this option. But today in this
session, we are going to experience the describe with AI. This is an option
which allow us to upload an image and
generate its description. Simply you will click
to this option. It will ask you to
select an image. Either you can select
from your generations, which you have been generated, or you can select
from a community. I will be picking from the
community at the time being. Let's see what creations
from the community we have. So I'm going to pick one of the creative art world
from the community. So let's pick this image, which is a featuristic city. A lot of buildings are around. So I have been selected the
image to generate a prom. So this AI have reaped the
uploaded image itself, and it's described it's
all the prompt, itself. So it's very brief.
Okay. So let's do the necessary configuration
to generate an image. So whatever simply, I will keep the settings and preset as
it is for the time being, and I will select the quality. Not a fast, I will select the quality as the
premium feature. I will keep the 16 by
nine ratio screen size and as large as possible. And I will keep
the four images to generate and then I will
keep as a private mold, so it will not be a
combinar recreation and PhotoL is also selected, and I will keep other features as it is for the time being, and I simply click
to the generate. This problem is automatic for us depending on the
uploaded image. So this is one of
the as features. And it can read whatever images, so you can try with your images, which you have either is being captured from by itself or AI. So it's being it's
consuming around 40 tokens. And it's a four images, so it's around ten tokens
for each pictures. If you want to modify the
count of images, so from here, you can do it or you want to do the different resolutions, so this is an option here, which you can do this differ
resolution itself as well. So this is an image have been generated and you can
see how similar it is. So the image which I
have been uploaded, it's keeping the same concept, but picture is totally different and is very high
resolution, by the way. So you can see the four variants of the picture which
I have been uploaded and keeping the same concept
or a context of the image. And it's one of the best
feature, I would say, when you are in a
ruh or you want to be that you think that you can you will not need to be
describe the image as it is. So this is one of
the best feature. I hope you like this feature
and you will try it. So see you in the
next lecture for the new or more
features. Thank you.
Anas Iqbal, Tech Mentor
