Transcripts
1. Introduction: Are you worried that AI will
replace you as a creative, or you've been intrigued and interested by the world of AI. But as soon as you
start looking into it, it feels quite overwhelming as you don't really know
where and how to start. These are some of
the reasons why I really wanted to
create this class. This class is not going
to be a learn from me, but more a learn with me. As I'm still learning and navigating this ever
changing world of AI, and I am by no
means an AI expert. So I just wanted to take you along and show you what
I've learned so far, and what are some of the things that you can achieve with AI. These are some of the
images that you'll be able to generate by
the end of this class. We'll create this
dreamy dropper image from scratch in mid journey, using text based prompts, and we will then integrate
our design in Photoshop. We'll also create another
image where we use product photography to
showcase our packaging design, and we'll enhance it
using Adobe Firefly, Photoshops AI, adding elements such as
bubbles and foam. Hi. My name is Beatrice Manis. I'm a branding and
packaging designer. I love helping product
based businesses bring their visions to life through branding and packaging, whether I'm working with
small businesses or entrepreneurs or freelancing
for international agencies. I love experimenting
with new softwares. I'm usually one of those
people that has to download the latest Adob Illustrator or Photoshop update as
soon as it comes out. Which is why I've been
doubling with AI, as I've been really
curious to see how it can help me in my creative
workflow and process, especially when it comes to packaging design and
product photography. As I haven't seen a lot
of that out there yet. In today's class, we'll touch
on different AI tools and how they can be used
at different stages of the design process. We will of course
keep a focus on packaging design and
product photography scenes. So these are some of the cases
where I've used AI so far. For example, at
the very beginning of a new project proposal, if I'm working on design
proposals and I want to help the client envision
this brand role we're creating for
their product, that's one way I've been using supporting
images using AI. Or at the very end
of the project, a lot of the times
you finish a project, and either you or the
client might not have the time or the budget to hire a photographer and have
a proper photoshoot. So I've been working
with AI in combination with photoshop to create
my own photography scenes. And last but not least, creating supporting images
just for social media. It just helped and took a lot of pressure off showing
up constantly, and it's made my
life so much easier. We'll kick off with M
journey and discord. I'll walk you through
what they are, how they look like and the
basics of how they work. We'll then explore a couple of different methods
of image creation, learning how to create
our own discord channel, and how to write
successful prompts. We'll also touch
on features like describe and imagine
with image references. Once we're happy
with our images, we'll take them into photoshop, where we will import
our packaging designs, either as mockups
or as photography, and seamlessly integrate them into our AI generated
compositions. Here we'll also use some of
the photoshop integrated AI, Adobe Firefly, which
is very handy. We'll learn how to add
or remove elements with generative fill to enhance
our image even further. My in for this class
is to constantly updated with new
lessons and new content as there's a lot
of new tools and new updates happening
in the AI world. Make sure you keep an eye on the class and when new
lessons will be added. This class is perfect
for designers at any level with a basic knowledge of Photoshop and Illustrator. Designers who are curious about new technologies and
tools for their workflow, and for those who want to level up and experiment
with their image creation. Let's get into it and
start learning together.
2. Class Projects: For the class project,
I love for you to experiment and play around
with any of these softwares. The aim of the class re is
to create one AI image. So it's up to you how
you want to do it. You could either create an
image from scratch using mid journey and integrating photoshop to add your designs, or if you're not ready to
create an image from scratch, you could open an existing
image or a mock up into photoshop and use the AI tools in Photoshop to add
or remove elements, expand the background, or just really be creative to
enhance the image. One thing that I would say
with either method is, be prepared to be patient. You can achieve really, really cool results with AI, but it takes a lot
of trial and error. So be ready to fail a few times and rewrite those prompts many, many times before getting there. But trust me, the results
will really be worth it. We'll create this
dreamy dropper image from scratch, mit journey, using text based prompts, and we will then integrate
our design in Photoshop. We'll also create another
image where we use product photography to
showcase our packaging design, and we'll enhance it
using Adobe Firefly, photoshops AI, adding elements
such as bubbles and foam. I can't wait to see
what you get up to.
3. Disclaimer: I feel I need to kick off this class with a big
thought disclaimer. As I previously mentioned, I am in no way an AI expert. I'm going to teach you what I've taught myself
through a lot of trial and error and failing a bunch of time. I'm
not going to lie. It's going to be a bit of
a steep learning curve, but just don't lose
patients and, you know, just persevere
because the results that you can have
are mind blowing, and it's been so
helpful integrating all these AI tools in my
day to day designing. So if you have
something in mind, and, you know, the first
time it doesn't work, the second time it doesn't work or the third time
it doesn't work, don't give up and
just keep trying, change a word, change a prompt, and trust me you'll
eventually get there. A second disclaimer, This
time is about copyright. And as you know, it's a bit of a gray area when
it comes to AI, so please bear with
me and make sure to read all the correct software
and tools guidelines. I've added them in
the references, so you can go ahead and
do your own research. But just to keep
it really simple. When you're creating
an image in journey, the image you create, you don't own the copyright, and the image is there
for the world to see, as it's all public domain. When it comes with Firefly, if you're creating
an image in Firefly, the Firefly model is trained on stock imagery or
open license imagery. On general rule
and this goes with mid journey, Firefly,
or anything. When you're using references, because you are able to image references to create
your own AI imagery, make sure you're only using images that you're
allowed to use. I will leave all
the necessary links in the project description for the mid Journey
community guidelines and the Adobe Firefly user
guidelines alongside, more AI tools, alternatives and websites you can
draw inspiration from. Let's get into it. In both scenarios,
and in both cases, as you will see in
the guidelines. Obviously, you're not allowed to use images as references
without the artist's consent. So for example, if I'm creating
an illustration, or well, if I'm using an image of an illustration as
an image reference, I would either use one of my friends illustrations and
ask for their permission, or if there's a specific one
that you really want to use, make sure to reach
out to the person and ask for their consent. That is always a very
important thing.
4. Midjourney & Discord: So what is Md journey? Mid Journey is a generative
AI software that produces high quality imagery from simple text based prompts. Mid Journey is just one of the many softwares out there like Tally and Sable diffusion. I've personally picked
M journey because it's the one I've been playing
and experimenting with, but you can feel free to
try any software out there. Unfortunately, Mid Journey comes with a little disclaimer. They used to offer free trials
when it first launched, but that's no longer the case. So if you want to play around
and create AI imagery, you're going to need
a subscription. The good thing is, they're
all monthly based, and they're around $10 a month. So if after a month you
decide AI is not for you, you can cancel at any time. The good thing about
M Journey as well, you don't need any other
software or hardware knowledge. The only other tool
that you'll need to be familiar
with is this Cord, which brings me to
the next topic. What is this Cord? This
Cord is a messaging up, just like Slack or WhatsApp. It was originally
created for gamers to discuss the
strategies online. You will need to create an
account, which is free, as that's the platform
that you'll be using to communicate with Met Journey
Bot and create AI imagery. Let's get started by going on to Met Journey's website
and click on Sign Up. As you can see here,
you have two options to either log in with
discord or with Google. Let's head over to
the discord website as we need to create
a new account. In this case, we can type
discord.com Flash Register. And here you'll be prompted
to create an account. Let's go ahead and fill in all the information
such as e mail, display name, user name, date of birth, and once you're all done,
click and continue. You'll of course have to
go through a Capta to prove that you're a human
and verify your account. Once you're all done, you'll be prompted to create your
first discord server. You have a bunch of
different options here. But let's close it
for now as we want to head back to Mid
Journey website. This time, we can
click on sign up, and now that we have
a discord account, we can click and
continue with Discord. As you will see,
you'll be prompted to accept an invitation from
the Mid Journey bot. Once you do, you'll be able to access the discord
M Journey channel. As you can see, there's a lot of rooms and a lot of different
channels and chats. Feel free to go ahead and
explore this a little bit. As you can see the Newbies
room are where you'll be able to see the stream of
imagery from other users. While you don't want to
necessarily use these to create your own imagery as
it can be quite distracting, they're great to
get inspired and learn how others write
and use prompts. If you head over
to Trial support, you can see any
questions and answers other users might have
had and out your own. Let's click on getting started. These are the first steps
that you'll need to do in order to create
your own images. Click on the first step, which will take you to the
mid Journey's website, and this is where you'll be
able to see your account and other finalized AI
images from other users. If you click on any image, you also be able to
see their prompts. Let's go ahead and log in now. Again, use this cord. And authorized access. Now that you've logged in,
you're able to click on some settings like picking
light mode or dark mode, and then you're able to click on your account name and click
on Managed subscription. As I mentioned previously, you do need a subscription to play around with Md Journey. You can go ahead and
pick any of these. The basic plan
allows you 3 hours, so that should be a
good starting point. See you in the next
lesson where we will learn how to create your
own discord server.
5. Creating a Discord Server: I'd highly recommend for you to create your own discord server. By doing that,
you'll pretty much have your own private channel
to work on your images. While if you're using one
of the main channels, you'll be distracted by a lot of outside noise and a
constant stream of images. Let's log into discord
and we were a minute ago. As you can see here, you
can click on ad server. Here you're prompted with a
lot of different options. If it's a specific server for gaming or for anything else. In our case, we're going to
click on create our own, and this is just for you for
now, so we can skip this. You can customize your server by adding your username
and an image. Once you're done,
click on Create. As you can see, now you've got your own server and the
main general channel. Now let's go back to
the New Bis channel until we find the
Md Journey bot, the one highlighted in green. Click on it and
click on Add App. Here you want to
select Add to Server. Here you have to
select the server which you want to add
the Mi journey but to. We can of course
select the one we've just created and
click and continue. By doing this, you're allowing
the Mid Journey bout to live in your server and
communicate with it once again. Go through all the checks
and once you're done, you're prompted to
go into your server. As you can see, the
Mid Journey but is now in your channel and
ready to take your prompts. If we click on the plus
next to the text channels, you can create more channels. You can skip all
the tips for now. For example, if you
want to work on a specific image for a
project or for a client, you can create a dedicated
channel just for it. For example, here, I'm going to create a channel
for basic prompts, and you don't need to
have it private as you are the only person in
the server at the moment. Just like that, you have to
create a different channels. See you in direct class
to create some images.
6. Basic Prompts: As previously mentioned, M journey works with
text based prompts. But what does that really mean? You would describe anything
you want M journey to create by using simple
words and natural language, like you were talking
to a friend or almost. In this lesson, we'll go through the anatomy of a basic
prompt and we'll just create something really
simple just to show you some of the words they will be using and where
to place them. Let's get into it.
Let's start by creating and writing
our first prompt. As you can see here, if
you type forward slash, a lot of different
options will pop up. Let's start by clicking
on the imagine, which is the one that you
would use for any prompt. Here you can see an
example of a basic prompt. It can be just a simple sentence
with very simple words. Now let's go and write
our very first prompt. Something very simple. White beer can on a white plinth on a white,
isolated background. Let's just hit enter and see what M journey
comes up with. As you'll see, M journey
creates four different options, Let's click on them
and see them all. As you can see here,
it has followed all of our input correctly and has
created what we've asked for. Here you have different options. U stands for upscale
and V for variant. Up scalers are used if you
want your image to be bigger. The variants are
there, let's say you like one of the images
that Md journey created, but you would like to
see a little variation from the selected image. The two arrows are just to mid journey to review
the same prompt. Now, let's say that we like
the second image so we can click on V two and see
what M journey comes up with. As you can see here, is taking the cylindrical plinth
that we liked and created four slightly
different variations. I really like the last one. So I'm going to click on U four to upscale
the fourth option. As you can see, the image is isolated and has a
higher resolution. You'd always want to
upscale a couple of times until you're happy with the size and the
resolution of the image. You have upscale creative, which makes it bigger, but also takes some
creative control and adds a few more
different elements, and upscale subtle, which just makes it bigger
in resolution. You also have very subtle, very strong, and very region. Let's just do upscale Sato for now to make
our image bigger. As you can see, our
image is now bigger. If you click on Opening Browser, you can actually see the size of the image and zoom into it. Now let's go back
into our channel. Let's go back to
the original image and click on V region. Here is where you're
able to select the specific area
and add any inputs. In the specific case, I'm selecting the whole
can as I want to add water droplets to make it a bit more realistic
and refreshing. I'm going to add
that into my prompt. Once I'm done,
click on the arrow. As you can see, M Journey has created four
different options, and one of them has even taken in consideration
the pink lighting, and that's reflected
on the droplets. Once again, when we're
happy with the image, we can upscale it a couple
of times and then save it. Just like that, we've
created our first image. Now that we've learned how to navigate and create
a basic prompt, let's move to the
next lesson where we'll be creating a
more detailed scene.
7. Image Creation: Let's dive straight into it. For this specific image, I want to create a very
dreamy and magical look for a skinkre brand with design branding and
packaging design for. Let's start by creating a
new channel for our image, and let's name it
dreamy dropper. Once you've created a channel, you can type forward slash and imagine to start
writing our prompt. As you can probably tell,
I've tested this before, so I'm going to go ahead
and write my prompt. I'm going to write
product photography because that's a photography
that we want to achieve. As an angle, I'm going
to write top town view, and then I will input the product that I want
my journey to imagine. A white transparent, semi frosted dropper skincare
bottle laying down against a bright sky blue
background with clouds surrounded by small
colorful wild flowers. Then I will add
something that helps us image this
magical atmosphere. Dreamy scene atmosphere,
sunlight refracting. I'm not going to add
anything else at this stage, and I'm going to press Send. As you can see, we have four
different options here, and it's already starting to look like something
I'm happy with. But let's dig deeper and try to amend the prompt and
see what we can get. Let's go ahead and
just copy the prompt and then type
forward slash again, select imaging, and
past to the prompt. I'm just going to remove
this comma here as I want my clouds to be integrated
with the background. As you can see by just removing
a comma, now they are. Something as simple as that
can make a huge difference. Now let's go again and copy everything and write
the prompt again, while making more changes. I'm going to actually
write something that will change the style
of the photography and write Fujicro Provia 100 F. This is a specific film as I want to achieve a
specific grainy look. Here you could write anything, could be 35 millimeter
photography, or if you like a
specific look of a specific film or
a specific camera, here's where you could write it. And then I'm going to hit Enter. As you can see,
it's a bit grainier and the colors are a
bit more saturated, which is the L I was looking for. I really like option two. I'm going to go
ahead and click on V two to see what
journey comes up with. Here's our four
different images. They're starting to really look like what I've envisioned. But I want the product to be
more of a angle this time. I'm going to go ahead
and copy the prompt, type forward slash and imaging and paste
the prompt again. Here I will replace the
angle with laying at a 45 degree angle to see if
mid Journey picks it up. Then once again, sum done,
I'm going to hit Enter. As you can see, now
the product is at a 45 degree angle,
which is great. I really like the second one. I'm going to hit V two and see what type of
variations we get. I have once again four
different variations. I actually really really like the first and
the second image. I'm going to go ahead
and upscale them both just so we can have two different
options and compare them. I'm going to hit on
upscale Sto and upscale creative to see how that changes in between
the two up scales. Don't worry if this
happens sometimes. Mid journey might be really
busy with a lot of requests, so you can just click on redo Upscale and Mid
journey will do it again. Here we have our image, and here we can really see the difference in
between the two. Upscale subtle is
really really subtle. It might just be the
detail of some flowers. We can just pick the
one that we like the most and open
it in the browser. Here you'll be able to see the level of detail and
how big it really is. Of course, you can
keep trying and slightly changing the
prompt until you're happy. Here's my final
result after changing a few parameters to get some more warmth and the
position of the dropper. As you can see here, this is the exact prompt
I've been using, so you can copy and paste it and use it as a starting point. For example, here you
can change anything. I would keep product photography if that's the look the
you're trying to achieve. Of course, you can change
the angle as we have done. We've started with the top
view angle and we asked me journey to achieve
a 45 degree angle. You can then replace the
dropper with any other product. It could be a shampoo bottle,
a chocolate packaging, anything that you're
currently working on and that it would
fit these aesthetic. Here I have used small
colorful wild flowers. But for example, you are maybe wanting to use white
daisies or maybe lavender, maybe you're working
on a fragrance that has lavender in it. You might want to showcase that. As you can probably imagine, you can spend hours and just keep changing and tweaking
until you're happy. Up next, we'll look at
a few other commands Image journey that can help
us with our image generation.
8. Midjourney Commands: Now that we've learned
about the imaging command, let's have a look
at other few ways we can interact with
them in journey aboutt. Command are there
to create images, change default settings, or perform any
other useful tasks. There's a lot of
commands and we'll only go through a few of them, but I will leave a list of all the ones that you can find
and how you can use them. Let's get started by using
the imaging prompt again, but this time using
image references. So let's click on upload a file, and I will just click on the files that I want
to use as references. In this case, the first image is a bottle laying in the water, and I'm going to press enter, and then I'm going to
upplod the second image, which is a pinkish water scene. You can also uppload them at the same time. It's
entirely up to you. I'm going to click
on Opening browser and copy the link of
one of the images. Then type forward
slash, imagine, paste the link of the first image and do the
same for the second image. Now M Journey knows that
whatever I'm asking it to imagine is going to have the vibe and look and
feel of these two images. I'm going to go on and
write skincare theorem, transparent glass bottle with white dropper, and hit Enter. As you can see here,
we have four results which have merged the
two imagery we gave it, which is really cool, and it's also pretty accurate
with what we've asked. I want to create something still magical like my first image. I will add dreamy serine
atmosphere, sunlight refracting, which are two of the
props that we had in the previous image when we were creating the main dreamy scene. I'm going to use the same
type of camera or film. I'm going to add Fugcro
100 F again and hit Enter. As you can see, M Journey has created four different images. Again, we have a more
sly vintage retro. I really like the first
one that he has created. I'm going to copy the
prompt and paste it. Once again, tie
forwards lash, imagine. Since we want the same product as we had in the other image, I'm going to copy
the same prompt. I'm going to write
white transparent semi frosted dropper
skincare bottle and delete the part
that is no longer relevant and hit Enter again. We have four options where
the dropper is metallic. I'm going to copy
your prompt and paste it again as I want
a white dropper. I'm going to write white
dropper and press enter again. As I mentioned, there's
going to be a lot of iteration like this where you can just copy paste the prompt, change a word, and
then hit enter again and see what Ma
journey comes up with. Now we're getting really close
to what I'm looking for. I really love the second image, so it's pretty perfect as it is, but I'm going to ask
Ma journey to do a few variations to see
what it comes up with. As you can see now has decided to pick a different
angle, which is really cool. It's now no longer top down
view, but it's standing. These two images could be
used together as they could belong to the same photoshoot,
which is really useful. Now let's up scale the first
one that we really liked, and we can also zoom
out 2% or 1.5%. What does that mean is
that mid journey will zoom out of the image and
fill the space around it. This is great if you're creating
images for social media, for example, and you need the
background to be expanded. As you can see, it has once again created four
different options. I can click on them
and it has extended the background keeping our
product in the center. The slight variation here
might be the reflection. It's a bit darkened on one
of them versus another one. Once it has generated, you can click on the image
and opening browser. Once again, you can zoom and see that the image
is much bigger, so you can go ahead
and save your image. Now let's go back
to the channel. There was another image
that I really liked, so I'm going to pick it
again and click on V four to see what different options mid
journey will give me. As you can see here, I have
four different options. What I'm going to do next is I'm going to pick
the one that I like the most and I'm going to up scale it just like
we've done before. As mentioned, feel free to upscale as many times as
you need and as you like, depending on the format and
resolution that you need. Just like that, we've created two images that have
the same setting. Once we import our own
packaging design in photoshop, they could be used
for a case study, for example, or for the
same social media post. Now that we've looked at
imagining image references, let's have a look at describe, which is great for
prompt inspiration. So here I'm going to create a channel for it and
type forward slash. As you can see, a lot
of options pop up. Let's click on the
scribe and here you can either choose
images or links. In this case, I'm
going to pick image and select the image that I
want to add and press enter. Don't worry too much
about the image here as this is just
to generate prompts. The final result will include some of the
elements in the image, but it will look nothing like the image that
you're uploading. You can upload any
type of imagery. As you can see, it has generated four options for four
different prompts. Let's go ahead and
imagine them all. A lot of the time if you're using an image with a product, M journey will pick up wards that are written
on the packaging, which we can of course
delete at a later stage. At this point, we
just want to try all the different prompts to see their different
look and feels. Once we find the
one that we like, we can of course refine the prompt and add
or remove words. As you can see, Mi
journey has created four different options and
we can click on all of them. Each set of images have all of the elements that we
saw in the original image, but they look nothing
like the original one. It has the bottle,
it has the hand, has the shower curtain, and has an overall
pink aesthetic, which is what the original image had and was the
La was going for. Now we can go through them and decide which one fits the
esthetic that we want the most. We can do different
variants, we can scale them. As we've done previously, we can copy and paste the prompt that created the closer
esthetic that we had in mind, and just go through
the same steps we've done in the
previous class. We can copy and paste
the prompt that created a closer esthetic
of what we had in mind, and just go through
the same steps we've done in the
previous lesson and trick the prompt again and again until we are happy
with the final result. Now that we've learned
a few more commands, it's time to save our
AI generated image and move on to Photoshop. See you in the next lesson.
9. Photoshop Editing: Now it's time to combine
the AI generated image with created image journey with the product and
packaging with designed. You can do this in a
couple of different ways. The first way could
be using a mockup. By using a mockup, you
would already have a warp to it and
highlight and shadows. When it comes to
merging the images, it will make your
life a bit easier because you will have
a base to start from. Or let's say you
don't have a mock up, but you have an image that
the client has shared, or you do have the
actual product, and you can just take
a picture of it. Both scenarios, always
keep in mind about the highlights and shadows
and the composition. You want to be able to match
the two as much as possible, especially if you're taking
a photo of the product, make sure you're
doing it after you've generated your composition
in mid journey. By doing that, you already
have an idea of where to position the product and how to use the highlights
and shadows. Let's start by opening
our image in photoshop. The first thing
that I always do is create a copy of the layer
and hide the original, so we can always go
back to our master if we make any mistakes
and we want to go back. The first thing that I'm
going to do is I'm going to actually use
Photoshops AI Firefly, as there are some
flowers that look a bit weird and I want
to delete them. I'm just going to click
on the patch tool, but you can click on
any selection tool, and this generative
fill bar will pop up. Click on the generative
fill and you can write remove or just hit enter, and Photoshop will automatically read the image and
replace your selection. As you can see, it has just
removed the flower entirely. We usually get three
different options, but in this case, it's
super straightforward, so any of them really works. It's really cool and quick, how you can do
something like that. When doing this, it creates
a layer with a mask. It's a non disruptive layer. That means it's not affecting our original image,
which is great. Now, let's repeat the same
process for a couple of more elements to
make sure that we have a bit of a
cleaner background. This tool is super helpful
and saves so much time. Instead of using the
clone stamp tool and just brushing in, we can with a
click, just replace whatever element was there with the background,
which is great. Once we're happy
with our image and it's looking cleaner and
how we want it to look, we can open our mock up. As you can see here, I'm
using an old mockup. It doesn't really
matter if the bottle is not the same because
in this case, we're just going
to copy and paste the label layer with the
highlights and shadows. Let's go ahead and paste it onto our original image and adjust it so it matches the
product that we have. Once everything is in its place, I like to group all
of my AI layers and color code them just to
make my file super clear. I'm then going to
pick my magnetic laser tool in this case, and I'm going to start tracing the bottle from the
original image. I'm doing this, so
I'm able to perfectly fit the label and just
isolate our product. Once I'm happy with
the selection, I'm going to copy the
layer and create a mask. Now I can apply the
product mask selection to our label to make sure it
fits within the boundaries. Cool. Now I'm going to
open our label group, double click on
the Smart object. Here is where I will go over to Illustrator and add our design, which I will paste into
photoshop as a smart object. Always remember to paste
this as a smart object, so you're able to go back into Illustrator and
make any changes. I'm going to adjust
it to a rough sizing, hit save, and close. As you will see, it will automatically adjust onto
our main photoshop file. Here we can go ahead and tweak the highlights and shadows of the mockup to actually match the highlights and shadows of the scene that we've created. But at least we have
a starting point. And we can tweak our
label, make it bigger, just to make sure that it fits nicely and naturally
on our product. Once we're all done, we
can start adding curves or different adjustment
layers until we're happy with the
overall look and feel. I'm adding some curves here to make sure that our
design stands out. I'm also going to add a hue saturation
adjustment layer and play again with colors
to make sure that they are fitting the
aesthetic I'm looking for. Once I'm happy with the image, I will select all the
layers and copy them. Whilst they're all selected, I'm going to merge
them into one layer. Once merged, I'm going to select filter, others, high pass. Here select a really low
value and hit Enter. Now select the great
image layer that has been generated and set the
transparency to overlay. As you can see, it just makes
all the details crispier. This is such a cool filter
that you can add on top of any image to make
it send out and pop out. Now that we use the
mockup method for the AI generated image that
we've created together. I'm going to take another
image I've previously created to show you how we can do the same thing but
with the product. In this case, I'm going to use product photography that
the client has shared. These were just neutral images against the
white background, then they've used
for the website. I'm going to show you how we can still match them together and make sure the high light and shadows and everything blends, so it looks seamless. Let's do the same thing
with the second image that I've previously
created in M Journey. For this image, I'm going to use a photograph from a client. I'm just going to import
the image and roughly place it on top of the
one that I've generated. As you can see, it's
not the same size, but it doesn't really
matter at this point as the original AI product
image won't be there. We can copy the
original layer and paste it above our packaging, selecting the mask
of our design, and click on the little chain
to unlink the two layers. We're going to go bigger and
distort the image just so the same texture
is going to fill the actual size of the product
that we're going to use. Don't worry too much here
as we're going to use it as lighter color and tweak the opacity to 50 and
the fill to 50 as well. As you can see, adding in some texture of the bubble
onto our actual label, integrates it a bit more
with the original image. Now let's have a look at the original product
generated in mid journey and
how the highlights and shadows are
affecting the product. As we want to replicate
the same highlights and shadows onto our product to
seamlessly integrate it. Let's start by creating a layer. I'm going to start by
creating a highlight, so I'm going to select a very light p just to match
the overall aesthetic. I'm going to click on
my pain packet tool, or it can be a brush
just to fill the layer. Once it's been filled, I'm going to turn it into a smart object, so we can add different
adjustment layers. I'm going to create a mask, so it fits within the
boundaries of our packaging. Then going to click
on soft light. As you can see, it's starting to add some dimension
to our product. I'm going to do the same
thing, but this time, I'm going to use the
original product image so that this one has all
the bubble reflections. Once again, I'm going
to do the same steps, converting it to a smart object, using the same mask, and setting the transparency
to soft light, and adjusting
opacity in the fill. This just adds another layer
of realism into our image. I'm then going to
create a new layer and repeat the same process for
the other side of the bottle. I'm going to create
a thinner highlight and I'm going to do the
same on the outside two. This time, it will be a
darker one as it's a shadow. I'm going to use darker pink
and set it to multiply. I'm going to do the same
for the cap because the shadow would hit the bottle and the cap at the same time. Once we're happy with
all the results, we can group all the layers and create a group called product. Now it's time to integrate some generative fill like we've done for the
dropper image. But this time, I want to
add some bubble bath foam. Let's click on the Lasso tool to highlight where we want to
add the bubble bath foam. The generative fill
bar will pop up and we can write bubble bath
Foam and it Enter. Photoshop will give us
three different options. They're all looking pretty good, but to be honest, I want
it to be more pink. I'm going to select the area again and
write the same prompt. As you can see here, there's
a little image icon, which is reference image. You can click on it
and you can choose any image for photoshop
to take inspiration from. In this case, I'm
choosing an image of previously generated
image journey that I didn't end up selecting, but had really cool foam. I'm going to hit generate,
and as you can see now, my foam is pink as it's
taking inspiration from the image that I used as a reference, which
is really cool. I'm going to do the
exact same thing to add more foam on the
rest of the image. As you can see the foam is
covering some of my text here. But as these are non
disruptive layers, I can go in and brush out what I don't
want to be covered. Here has generated a bit of
a weird bubble on my foam. I'm going to select it
with my Lasso tool or any selection tool and just click enter in the
generative fill bar. I can either write, remove or just hit generate
and Photoshop will automatically give me
three different options that perfectly cover the area. As you can see,
it's even keeping the shadow that is
meant to be there. Now to make it a little
bit more realistic, I'm going to add
a little shadow. I'm going to create a new
layer and just brush in with a darker pink and
set it to multiply. Let's then create a mask so that the shadow will go
underneath the foam. I really want these
bubbles to be on top of my product to be even more
integrated into the image. What I'm going to do is I'm
going to create a copy of my original image and place
it above all my layers. I'm then going to select
the two bubbles I want to overlay the
product and create a mask. Then with my brush tool, I'm going to go in and just delete the parts that
are not relevant. One last thing that I want to do is I want to
adjust the opacity because we do want to
see some of the bottle behind the bubbles to
make it more realistic, as that's how we would look in real life as the bubbles
are transparent. Last but not least, I want to change the color a
little bit more. Make it a bit colder
and a bit more purple. We're going to do that
with an adjustment layer using hue saturation. Once I've applied it on top of everything, as you can see, my product is also
changing color, and we don't want
that to happen. I'm going to go into the product group and
select the mask and apply the same mask onto
my hue saturation layer. Once the product is selected, I can just brush off
the adjustment layer just for that part of the image. As you can see, my packaging is back to the original color, whilst the rest of the
image has my selected hue. Like that, we've
created an image I'm really happy with that was originally generated
made journey and enhanced in
photoshop using Firefly. As I've done with
my previous image, I'm going to select all of my layers and copy them
and merge them together. Then click onto filter, other high pass, select
a ide low number, and set it to overlay. As you can see here, the difference between
the before and after and how all the
details are crispier. We did it. And just like that, we've created a high quality AI generated image for
one of our designs.
10. Final Thoughts: Now that you have a
really basic knowledge on how to generate images in mid journey and how to enhance any images in photoshop
using Firefly, you can create as many
images as you like. The only limit will be
your own imagination. As previously
mentioned this class is a work in progress.
What does that mean? So that means that
whenever I'm going to play and experiment
with new AI and tools, I'm going to be adding lessons. So make sure to keep an
eye out to be notified. We kicked up the
class by learning about mid journey and discord. We then created our very first discord
channel where we've chatted with the Mid
journey butt and experimented with our
very first prompts. We then had a look at a few
more mid journey commands like describe and imagine
with image references. Up next, we created a more
complex image recreating a dreamy magical scene for a skincare product before
moving into photoshop, where we've integrated
the actual design, using a mockup or
product photography. Last but at Last, we played with a Wi Fi Fly Photoshops
Integrated AI to enhance our AI generated
image even further. It would be really great if you could share any images
that you create and you come up
with alongside with any roadblocks that
you might have faced, and any types of
tricks that you've learned along the way
in the project gallery. That way we can all
learn from each other. Last but not least,
since I'm aware that this is still a tricky and potentially overwhelming topic, I've added one to one sessions onto my main skill
share profile. You can use them to
pick my brain one to one or go through any of
the steps of the process, or even better, we can create
your own AI image together. As always, if you
have any questions, comments, or concerns, please feel free
to add them in the discussion tub or just
send me a message. Thanks again for taking the
class and see you soon.
Beatrice Menis, Branding & Packaging Designer
