Transcripts
1. Introduction: Hi, I'm Theo. Welcome to the
Vizcom Fundamentals, a short course designed to help you quickly become
confident in using Vizcom to generate new ideas as part of your design workflow. We will start by looking at all the tools
available in Vizcom. I will also show you
how to solve some of the common issues
faced with using Vizcom and how to export your
ideas once you're finished. Let's get started with
navigating the UI.
2. Studio Mode: When you first log into Vizcom, you'll arrive in the
file management page. This is where all your files are stored and where you
can create new files. You can create a new
file by clicking Create New File in
the top right corner. You'll see that there
are two options start in studio and
start in Workbench. The studio is quite
like Photoshop. It's a canvas that
you can draw on, generate on, and work
within a very familiar way. Workbench, on the other hand, is more like a
virtual whiteboard, and it's designed for iterating over multiple different images. We're going to
start off in studio and make our way to
Workbench later. When you arrive in studio, you will notice that
the main part of your screen is taken
up by the canvas. This canvas, just like
any other software is where you can
sketch your ideas. On the left side, you can
see the layer management. This is where we can create
new layers, rearrange them, change the aspect
ratio of the canvas, and also change the
background color. Next to layers, you'll
see the assets tab. Assets are things like images, colour palettes and prompts that you want to save to
reuse again and again. Up at the top, you can see all the main tools that we use to interact with studio mode. The first one is insert. This is where you can upload
images, three D models, and Brown's preset
three D models that we can bring
into our Canvas. Next, you have the
selection tools. The little drop down
shows you that you can swap between transform and warp. The brush, just like any other software,
is how you sketch. Make sure you're on a sketch
layer and not a model layer, and you can sketch
directly onto the canvas. Next to it is the eraser. There are two erasers inside the studio mode,
the normal eraser, and the magic eraser, which tries to use
the context of what's around the thing you're
erasing to fill it in. Next, you have the colors. And this is how you choose what color your sketching with. We also have presets
for rectangles, ellipses, and lines that we can draw directly
onto the canvas. There are multiple different selection modes
available in studio, including the asso, rectangle, brush, Bezier, and
auto selection. These allow you to select
different parts of the canvas. And after selected, you can
adjust with the offset tool. There is also a symmetry mode. This will be graded out unless
you're in the brush mode, and when you click
it, provides you the ability to draw
symmetrical clean shapes. The workbench button
allows us to swap between studio and Workbench
mode whenever we want. You'll notice you've got
a floating render window. When we actually go and use
the AI tools inside Fiscal, this is where the options that
are generated will appear. It's here, you can select
which one you want to view, and whichever one
you want to choose, you can click Apply. Over on the right, you can
see we have the modifier tab. This is where we can put
text based prompts into Vizcom to change what
we've drawn on our Canvas. We have some presets available. And if you were to click change, it would look at
what's on your screen and give you suggestions of things you might want to change. We also have the ability
to use instant render, new view, and variate
form and color. And we're going to cover
these in more detail later. On the side tab, you'll
see that there are a couple more options
for various AI tools. Variation gives you this
two axis graph where you can change between
different attributes to manipulate what's on screen. The colour palette detects
what colors are on screen and allows you
to swap them out. Make three D allows you to give the AI whatever is on your screen and ask it to
create a three D model from it. You can choose between low poly balanced and
highest detail, and you can also generate standard and HD
textures as well. At the bottom above
the generate button, it will give you
an estimated time depending on what
you've asked from it. Advanced options like
part separation, give the AI the opportunity
to split the model up into different parts based on material or based on them
actually being connected. The create tool
is a legacy tool, so you're not going to
cover it as it's been replaced with more
competent tools. Finally, we have
the adjust panel, which has some AI tools like enhance and remove background, but also more standard
image editing tools like exposure or Hue.
3. Workbench Mode: Let's have a look around
the UI workbench mode and see how we navigate it. In the top left, we
have the Vizcom button, which allows us to go
back to our project menu, and also gives us
various options that are very familiar from other
pieces of software, such as creating a new file, undoing and redoing,
zooming in and out, and also changing from
light to dark mode. In the top at the
center, you can see, just like in studio mode, we have all the options for how we're interacting
with our images. Insert gives us the
opportunity to insert images, both from our computer
and from our phone. And to insert from your phone, you simply scan the QR code with your phone and
upload your images there. All of these options
are different nodes that allow you to modify
and iterate on your ideas. And we're going to go through
each of these in detail. Your asset library, just like in studio mode, can
be accessed here. And the next set of tools
are all to do with how you organize and lay
out your workspace. As I said, it can
get very easy for this to become crowded
and a bit overwhelming. So it's really important to
stay on top of the layout. The text mode, as you'd expect, allows you to add a text in this makes it really easy to see
where you were doing different things and how the design has
progressed over time. Sticky notes I often use for
feedback and suggestions, such as saying that
this needs more shape. Sections allow you to break up this large canvas
into smaller areas, and you can drag
various elements into those sections
to create subgroups. This again, makes
it easy to group together different
stages of the process. So when you're going
through it at the end or midway through and you
want to review your ideas, you can see where different
ideas are grouped together. The Smart Dropper is a
really powerful tool, and we're actually going to look at this in its own video. Effectively, it allows
you to take style, color, and environment assets from a specific image and apply
them to another one. You also have undo and redo
buttons on this top tab. Over on the right, you can see we have our option to Zoom, the ability to export
various images. And the ability to share
the file with other people. You also have access to your account settings up
in the top right. In the bottom right, you'll see the question mark, and
by clicking on it, you can see various
locations that you can access different
documentation, tutorials and support.
4. Rendering Sketches: As you can see, I've imported
a sketch of a mocha part, and I want to start exploring different surface
treatments and colors. We're going to go
up to the top and we're going to click Workbench. As you can see, it's taken us
out into this infin Canvas. If you've ever used a
node based piece of software like the shader or
geometry nodes in Blender, this is going to
feel very familiar. You have the input, which
is your sketch or image, and you're able to
drag out the side of it and choose from the
various nodes available. Modify is the same as we were
looking at in studio mode, and allows you to
Instant render, generate new views,
variate form and color. But we also have a
couple of extra options like change expression
and extract. We're going to start
with Instant render. And as I said before,
the target fidelity, you can control what kind of level of realism
you're going for. So for now, I'm just going
to go with color blocking, and I haven't got a
reference I'd like, so I'm going to leave it empty. I'm going to click on the
one and I'm going to change it to generate four
images at a time, and I'm going to hit Render. As you can see, Vizcom has
generated four images based on my initial sketch with just a very basic
amount of colour. Now, something I overlooked
was that the colour of the background of my sketch
paper isn't actually white. And therefore, on three
of these generations, it's kept that difference between the canvas
and the sketch. As you can see, it's generated some sketches
with the color, very basic color,
and a little bit of variation based on where it can imagine light and
shadow to fall. At the moment, because it's
the only one that hasn't got this variation between
the sketch and the canvas, I'm going to go with this
final one over here. Now, what I like to do when
I'm editing and generating a Vizcom is copy the one that I want to use down a layer so I can keep the outputs next to
where they were generated. And this just gives
me a bit more flow. Some people like to drag directly out of
that final image, but I feel like that gets
very confusing very quickly.
5. Developing Your Idea: We've got this mockup here with a very small
amount of shape to it, and I'm starting to think about how we can actually
change the design. Traditionally, mocha
parts use straight edges, and I'm thinking it'd be quite
interesting to see what it looked like if it has
a more rounded form. So I'm going to
click on the image, drag out of it, and click
on what are you creating. This opens the main text
based modification tool where you can actually
just describe in plain English what
you'd like to change. So I'm going to say, make
the mockup more curvy. In the bottom of this area, you can see there are options
for actually improving the prompt or for
transcribing it with voice. If I was to click
Improve prompt, it's going to read
what I've put and try and make it so Vizcom
adheres to it more carefully. So it's going to say, make
the mocker pot in at one, I this image, more curvy with rounded edges
and a softer silhouette. I can then choose to
keep or dismiss this. Just like in studio mode, you have presets of what
you'd like to change, and by clicking on them, you'll see the suggestions it makes. You also have the
option to change the aspect ratio of the image. By default, it will render the new images in the same
aspect ratio as the input, but you can change it
to various presets or to a custom aspect ratio. Finally, you have the
mode for standard or P. I'm going to
leave it on standard, and I'm going to
click four times to generate four new ideas. Anytime stuff is being generated
outside of your viewpot, it's going to let you
know at the bottom. As you can see, it started taking what I've put
in that text prompt, taking the initial image, and trying to combine them
together in various degrees. So the first one, it
softened the lines, not necessarily
made them rounder, but given it a little bit more curviness and a bit more shape. It goes all the way up to the extreme over here
where it really, really accentuates
the curves and actually gets rid of some of the stronger lines
that were on there. This is more of a middle ground
as it keeps those lines, but also takes some of that curvy shape and makes
it a lot more organic.
6. Using the 'Variate Form' Tool: Another way we
could approach this is by using the Variate tool. So I'm going to click and drag out of this initial
image as well, and I'm going to click Variate. Now, I know that these presets may work brilliantly for what
I'd like to achieve, but I actually want to use
the custom option because I want to control whether the surfaces or the
lines are curvy. So I'm going to click
on the top example one, and I'm going to type
in curved lines. As the opposite option, I'm going to choose
straight lines. And then for example
three and four, I'm going to swap them
out but for surfaces. So I'm going to do flat
surfaces and curved surfaces. Now, I'd like to see what this looks like to various degrees. So I'm going to go for curved
lines and curved surfaces. Straight lines and
curved surfaces, straight lines and
flat surfaces, and flat surfaces and curved lines to see
all four extremes. Now, it may be that I actually prefer the options
that are up here, but it's always helpful to see how different tools inside
Vizcom approach the same task. As you can see, these two
are visually quite similar. But even though this was
my favorite up here, I actually think this
one looks a lot nicer. Now, sometimes it's
going to generate images that are quite
similar to your input image. And that's just the nature
of working with AI. It can be a bit unpredictable. This one here I really like, as I think it's a
perfect combination of curvy yet still structured, and it feels feasible like
it could actually exist. So I'm going to copy this one, and I'm going to paste it down below just to keep my
file nice and neat.
7. Exploring Colour, Material & Finish: Now, I want to
start having a look at the colors because
at the moment, it's all silver, and I'm not really sure
that's what I'd like. So what I'm going
to do is I'm going to actually open this in studio, and I'm going to paint in where I'd like some different colour. So I'm going to go
up into the top left and I'm going
to take a new layer, and I'm going to go and
select a middle gray. To click off this. I'm just
going to very quickly paint in where I'd like slightly
different colour. I'm going to flood fill in the center as a very easy way
of filling all of that in. It doesn't have to be perfect. Vizcom going to fill
in the gaps for us. I'd also like this area to be a slightly
different colour. And if we're getting rid of
detail by painting over it, we can always combine those
images back together anyway. For this centre bit, I'd quite like it to be a
glossy chrome colour. So I'm going to do
it in white for now. Now, I'm holding down shift
to draw straight lines just to make my life
a bit easier at the moment because it
doesn't need to be perfect. And I'm going to flood
fill that as well. Now, that's roughly the colour
breakup I'm looking for. So I'm going to go back
out into workbench. I'm going to drag
out. And again, we're going to try this
in two different ways. Firstly, I'm going to try
it in a text based way. And secondly, I'm going to
use the Variate color tool. So with the text based prompt, I'm going to click on,
What are you creating. And in my describe
my changes area, I'm going to actually
just outline which colour I'd like
to change and to what. So I'm going to say, make
the gray areas silver metal. The black areas black plastic
and the white areas Chrome. And again, I'm going to click on the Improve prompt tool
and see what it suggests. It's going to suggest make
the gray areas brush silver, metal with a subtle sheen, the black areas matt black plastic with a
soft touch finish, and the white areas polished chrome with a mirror
like reflectivity. This is giving Vizcom
a lot more information to work with. So I'm
going to accept this. I'm then going to generate four images just like I did before. As you can see, it's now gone
and generated these ideas, and it's stuck really
closely to what I've been saying in the text area
and with my initial image. You can see this one
didn't respect that I wanted the spout to
be the black plastic, but the rest of it
looks really good. And if I wanted to move
forward with this one, I could always go into a
new prompt and tell it to make the spout the same
black plastic as the handle. The second image is a lot
closer and actually adhered perfectly to what I asked of
it, as did the third one, but the third one still
has some sketch artifacts, and they decided to put the release valve for the pressure in gold,
which is a bit weird. The final one is
still a bit sketchy, and while I like it, it wasn't exactly what
I was going for. Remember, if you generate all these images and there aren't actually any you'd like, you can either go back
and edit your prompt, or you can just
generate some more and just wait for it to give you
an option that you like. But for me, I really
like this option. So I'm going to copy it and
I'm going to paste it below.
8. Generating New Views: Now, I want to see this
object from a couple of new viewpoints, and
that's really easy. I'm going to click and drag
out from the plus button, and I'm going to
click on Modify. This again gives me the
option for new view. Now, from here, I
can select one of the many different preset views or the ones from the back, I can click Select a view to see the rear views and
the bottom views. I'm going to say, I'd like
to see it from the top, and I'm going to generate
a couple of images. And then I'm also
going to see it from a front left
three quarter view. Now, new view can be a little bit tricky to get
working properly, and sometimes it's
important to take a couple of different views and
combine them together. However, you can see that the top view has worked quite well, giving us a good understanding of what it looks
like from the top. Actually, all of them
have worked quite well, and these are great views, even though the
black plastic area around the spot has
disappeared again. But as I mentioned, you can normally go back and make that back to how you want quite easily with just a
simple text prompt. But these different views are really good because they give us good foundations for when we
want to animate later on.
9. Incorporating People: Now, I'm quite happy with
how this looks for now. So we're going to have a
look at how we generate images with people in and
also some concept renders. Putting people in a scene
really give a sense of scale and give a sense of and
where the product is used. So I'm going to
take this and I'm going to place it
down here again, and I'm going to click
the plus and drag it off. I'm going to click
the Try on button. And this, even though it sounds like it's
designed for clothes, also works really
well for products. Now, as you can see, you can add eight different items into this, but we only actually
need the one, and it's already been
filled in for us. If we have a specific
model we'd like to use, you can add them in here, but
you don't actually need to. In the pose, it gives
you a couple of presets for what kind of pose the person is
going to be in. But custom is always
the one I use, as it actually allows
you to dictate a bit more about the model
that's being used, just like if you had an image. So here is where you could
insert whether you'd like a man or a woman
to be holding it, whether you what you'd
like them to be wearing, what kind of pose you'd
like them to be in, and really kind of spell out
exactly what you'd like. So I'm going to
say I'd like a man holding the coffee pot. White studio. And just spell out exactly
what I'd like from the image. I'm going
to generate four. You may notice
that I pretty much always generate four images, and that's because I never
like to just generate one. I like to see a couple
of different variations of how Vizcom
approaches the project. And four normally gives me enough variation and one
that I really, really like. As you can see, it's generated four images here
of people holding the coffee pot in different
posers and different outfits. However, if you think
that any of these are really good but need
some minor tweaking, you can obviously take
the one that you like, paste it down here again, and then apply a
text base prompt. So I'm going to say make
the coffee pot smaller. As you can see, it's
generated some options, and it's actually in some ways kept the pose exactly the same, which can lead to it
looking a little bit weird. But that does look a bit closer to the scale that
we'd be looking for. So I'd probably be
asking it to regenerate these images with the hand
attached to the bottom. But that's how you actually generate things with people in. You can also inhere, instead of asking to make the
coffee pot smaller, you can ask to change
the appearance of the person or the
environment they're in.
10. Changing Expressions: Now if you wanted to control the expression of the
person in the photo, we have a specific
tool for that as well. So I'm going to take
the original image, and I'm going to drag it out. And in modify, I'm going to
click Change expression. Now, even though it would
be quite weird to have a specific expression of a studio model
holding a coffee pot, just for reference, I'll
show the kind of variation. So we're going to go and we're going to generate
an image that's sad and maybe an
image that's angry, just to see how Vizcom
approaches that and how it changes the
models expression. There we go. So depending on what expression you'd
like the model to have, the change Expression tool is a brilliant alternative to
typing it out yourself, and it gives you
that wide variety and that ability to really, really control how you
want the image to look, how you want the model to look, and how you want that
product to be shown.
11. Rendering In Environments: So let's have a
look at generating some renders in environments. There are two main
ways to approach this, either using the
text based prompt to tell it what kind of
environment you like, or to use a photograph of an
environment as a reference. And when you're using a
photograph of an environment, there are again two
ways of doing that, either by inputting it as a reference image or by
using the Smart Dropper. So I found this image
that I'm going to paste in of a kitchen
on Pinterest, and I really like
the aesthetic of it. So we're going to
try incorporating the mocha pot into
this environment. So I'm going to go
back up to here and take the image of
the product on its own and I'm going to paste
it down next to this image. Now to work with
multiple images in the text based prompt
section, it's really easy. All you need to do is
drag and select over both of them, drag
out of the plus. And when you click on
the text Base prompt, you'll see both of them are in here and they've both
been assigned a number. This is how you can tell Vizcom specifically what you want
to do with each image. Take a coffee part from at two. As you can see, when
you click space, the at two goes blue, and that means it's
recognized you're talking about this image and place it in the
environment from at one. Now, again, I'm going to click the Improved prompt tool
and see what it suggests. So it's being more
specific that it's taking the brushed stainless steel
mocker pop and placing it on the wooden countertop
in the environment from one with herbs and lemons
visible in the foreground. I don't want to make it
visible in the foreground, so I'm going to get
rid of that last book. I'm then going to change
the aspect ratio, and I'm going to
change it to square. And we're going to generate
four images like usual. As you can see, it's
generated four images, and there's actually
only one here at the moment that's any good. This one, the fourth one looks really badly
photoshopped on, and one and three
haven't actually respected the chrome
element or the shape. The only one that's got
close is the second one. But again, it's got rid
of the chrome band, and it's got rid of the
black plastic spout. So what I'm going to do is I'm going to try and combine
these two images.
12. Troubleshooting Render Issues: So I'm going to take
the one that I like, and I'm going to
place it next to the reference image.
I'm going to drag out. And I'm going to say,
make the mocker part into look like the
mocker part in at one. Now, it may suggest that I actually don't want
to use that language. And it's actually suggested
there to say octagonal base, fluted body, and black
handle n. I'm going to also add in spout so it specifically
relates to that spout. I'm also going to go
in. I'm going to add in that I also wanted to
use the Chrome band. And we're going to
generate some images. Hopefully, this will get us a little bit closer to
what we're looking for. And the thing with Vizcom
is, you're not going for massive changes in one step. It's all about small
refinements at this stage. As you can see, it's having some trouble keeping
to what we've asked. So I'm going to try
different prompt here. I'm going to say
replace the maker in two with the maker one. I'm just going to see
what it generates. And again, we're looking
for small changes and just refining at this stage. Now, again, this hasn't
generated what we'd like. So I'm going to do
is I'm going to choose the one that
is the closest. And all of these seem
to be fairly similar, so I'm just going to
take the first one, paste it over here, and I'm going to go in
and do a prompt. I'm going to say,
make the center band on the moker part
reflective, polished chrome. And I'm going to ask it
for a suggestion again. I'm going to accept it
and generate four images. Okay. Actually can see what it started
doing is actually making all of it reflective, and it's only the centre
bit I want reflective. So I'm actually going to go and I'm going to see
if I can mask out a specific area in studio
and bring it back in. So I'm going to take this image, I'm going to paste
it down here again, and I'm going to go
into the studio mode. I want to start telling
it specifically where to apply chrome and specifically
where to apply black. So I'm going to add a new layer. I'm going to choose
a bright red. I'm going to change
it to the brush mode. I'm now going to actually paint
on specifically where I'd like the chrome and
fill in the center. I'm I'm going to
change the colour to a different bright colour
like bright green. Actually, no, not
bright green because it's right next to this
green. I'm gonna go blue. I'm going to go into
brush mode and paint over and then fill
in the center. What I'm doing here
is I'm spelling out specifically to sc on where I
want to make those changes. I'm going to go back
out to the workbench, and I'm going to go into
the prompt section, I'm going to say, make the red band polished
chrome and the blue spout. Black plaster. Again, I'm going to ask
it for its suggestion, just to see whether
it suggests anything. And I'm going to accept that. I'm going to see
whether this works. Sometimes you actually need to bring in this image, as well, so it knows what it
looks like before, but sometimes it can figure it out like it out on its own. And there we go.
This final image here is actually started to respect what we
were asking for. It's given us
chrome on the band. It's given us black
on the spout, but it's kept the other areas
this brushed metal finish. And that's what you're
really aiming for. You're aiming to get
marginal returns each time and slowly get towards
what you're looking for.
13. Image Adjustments: Okay. I'm now going
to take this image, and I'm going to
go and edit it in studio mode. I'm
going to click Edit. I'd like to just make this
section a bit darker. So I'm going to go
and I'm going to choose the auto selection. I'm going to wait a
second for it to load. I'm going to click to select
that top section there. What I'm going to
do is I'm going to copy selection to layer, so it's created a new layer
with just that area on it. And that means I can now apply these settings
specifically to this area. And all I'd like to do is just darken it down a little bit, just to make it fit
a little bit better. That's all I really wanted
to change with that image, just to make that look
like it's made of the same material as
these other areas. But this is the kind of
render that you can use to show off what your product is
going to look like in situ, and it can be really
helpful in terms of creating that look and
feel and selling the idea.
14. Extracting From Images: Now, it may be that you have
an image with some colour that you like that you'd like
to apply to your design, and it's really easy to do that. I'm going to paste in
an image I found on Pinterest of these
ceramic marks, and they use four
really nice colour. There are two ways you can
easily apply the colour. The first one is to select
both and go into the modified. When you describe your changes, you can specifically ask for the colors from two
to be applied to one. However, another way of approaching this is to
use the Smart Dropper. So if you go up to the top,
you'll see that you have the Smart Dropper tool
and you can click the drop down to choose
what you'd like to analyze. If you're analyzing style, that means it's
the kind of image, whether it's a photograph, a marker, render, a sketch. Whatever you select with that will be applied to your image. Color is taking the colors of the product or the image
and applying them to yours. And the environment is
taking the environment, whether it's in a kitchen, whether it's in a field,
whether it's on a countertop. And you can apply that
to your product as well. We're going to choose
color, and I'm going to sample the
colour from this image. It'll take a second to
load, but once it does, you can see it selected those four colors out of there as being the
most important. Now I can go over
to the image of my product and it'll
say apply color. When I click, a new image
is going to generate, and this image is
going to try and take the colors from here
and apply them here. As you can see, what
it's done is it's tried to combine some of the
colour from here with here, and it's chosen to take this
brown effect and put that on the brushed metal and take this green and put it on the pressure valve and
this little hinge here. Now, obviously, it's really easy to variate a lot of
options with this, and you can take this and apply it as many
times as you'd like. This time, you can
see that it's applied the green to the spout instead. And it may be that you don't actually like the output here, but with enough
different generations, you should be able to
get something that applies in the way
that you're expecting. If you're not doing,
you can always revert back to just
selecting both and spelling out that you'd
like to use the green and the blue instead of the
chrome and the silver. As you can see,
it's not actually copied the blue
very well at all. So we may want to try again by just seeing
what it has to say. I'm going to change the
prompt ever so slightly. As you can see, the suggestion here has kept it quite close. It's not exact, but
it's quite close, and it gives us another idea. What we can do is
we can take this and we can go back to
our environment here, and we can actually swap the
colour out on here so we can say in the create so
in the modified tool, we can replace the maker part in one with the maker part two, improve the prompt.
And collect generate. And as you can see, it started
generating the same image, but with a new product in it. And this makes it really easy to stack the images
next to each other and show the different
colour variations that you're experimenting with.
15. The 'Smart Dropper' Tool: Let's try using a different
aspect of the Smart Dropper. Can use the style tool instead. So I found this render
on Pinterest that I really like of this
marker render of a watch. And what I'm going to do
is I'm going to select the Smart Dropper
and choose style. And it's going to absorb
the style of this image, and you can then apply that
onto your own product, and it's going to
generate a new image. As you can see, it's generated
an image of our product, but using the style as a
reference from this image. And this is great if you
want to come up with some concept ideas and things you're going
to go and take forward into Photoshop later. The final thing that we
want to look at with the Smart Dropper is
taking the environment. So we were looking before at how we took this environment
that we found a photo of on Pinterest and combined it together with a mocha pop
to create a concept render. Now, if we don't want to
necessarily use this exact view, but we just want to
take the environment, this inspiration, we can
do that using this tool. So I'm going to click
on the Smart Dropper and I'm going to
click on Environment. I'm going to go up to this
original image that we found, and I'm going to click
Absorb Environment. I'm then going to go down,
and I'm going to click on this image and see
what we generate. It's going to generate
the new image over here. Drag it down out the way. Now, what happens when
the Smart Dropper, I'll often not keep
the same perspective as the original image, and it will just use
the data as more of a general suggestion of what kind of environment
to place it in. As you can see, it's
taken that environment in a very loose sense
and applied it to the image to create a completely
different kind of image. As with all AI, every
time you generate this, it's going to give you
something slightly different. So if this isn't exactly what you're looking
for, by all means, just run it a couple
of times and see if it gives you something closer
to what you're looking for. But that's the Smart drop
it all in a nutshell.
16. Progress Review: Let's have a look at
this process overall, then and see how
we've developed. We're going to scroll
all the way back up to the top to this initial
sketch that we started with and adding a small amount
of color and form to it. You can then see how we use the modify and the
variation tools to go about changing the
style of the design to see whether we could get it closer to what we imagined. Then you can see how we went into studio mode and blocked out a specific area so the AI had a better idea of where we
wanted it to apply changes. Then we had to look at some new views to try and get it in more different perspectives to get a better idea of
the overall design. Then we had to look
at how we could put people into photos and generate ideas of in context renders and to see what the product would
look like actually in use. Then you can see how
we try to modify that to make it closer
to what we were expecting and look at
changing the expression of the person to fit the context of whatever the scenario is. Then we had a look
at importing in image and combining it with our product to get different outputs in terms
of in context renders. You can then see how we tried to spell out to the AI
specifically what we want to take from each image and
how it can sometimes take a couple of different
approaches to get it to do exactly
what you'd like to do. But eventually, over here, we got an image
that looks really convincing and shows the
marker part in context, and it gives a really clear idea of our expectations
for the product. We can then see how we took
the colour and materials from a reference image and use
them to generate new ideas. We also then looked at
how you could apply different styles and different environments
using the Smart Dropper. And after we generated
these blue ceramic ideas, going back to our
original image that we generated and swapping it
out for our new product. This is how we
create a nice suite of ideas and a suite of concept imaging that all feels visually related
to one another.
17. My Process: Over here, you can see when I
was initially experimenting with this mocha pot idea
before recording this course. I just want to show you how it worked there as well
because I want to give you a better idea of how
repeatable this process is. So I've split it up into
sections over here. And as you can see, I
took this initial sketch, and I gave it a reference
image of this piece of industrial design concept
to generate this idea. I then used the variation
tool and looked at swapping out the colors for something
slightly different. As you can see, I took
the pressure valve from the front and asked for it to be moved to
the side instead. Then I started looking at the variation again in
terms of form, and again, use the curved and
straight lines, flat and curved
surfaces to create some different ideas of how our surfaces actually function. You can see here how I was experimenting with
different colour, material and finish, and it's really easy to combine
different ideas together. So I really liked
this dark navy, and I really liked how this had gone a nice reflective gold. So I asked it to combine the blue elements of
the first image and the gold elements
of the second image to generate this final idea. From this piece of concept here, I went and instant rendered it to create a more
realistic image. This realistic image, I input it with an environment I liked and a perspective
that I liked to see whether I could
generate an image that was similar
to both of these. I got some different outputs that were different blends
of these different images. As well as this, I experimented with how we could
use an image of an existing product
similar to our own to actually generate
new images for us. So because this mockup
is sat on the Hb, it makes it really
easy for the AI to understand the perspective. So I asked it to fix
the perspective, and it took the image of our
mocker pot on the hub and did change the perspective to make it look more realistic. I then wanted to experiment with seeing the product
somewhat in use, so I asked it to add
some steam coming out of the spout to make it look
a bit more realistic. As you can see, we
ended up generating a really realistic piece of
concept art for the product, as well as a nice, more
open contextual render.
18. Bonus: Exploded Views: Now, I just wanted to cover
a little added bonus, which is the ability to
create exploded views, such as the one here. Now, how I did this was actually using a preset prompt that is delivered directly in Vizcom and it can be found
in the Assets panel. If you click on VSCO, it will give you
a load of options that are pre shipped
with Vizcom, one of them being
a set of prompts. All of these prompts
you can click on and you can actually copy them to use directly in
your modified in this case, I've chosen the exploded
view prompt and put it in alongside a preset image that is delivered with Vizcom
for training purposes. As you can see, they're a
little bit hit and miss. But this one here actually
looks really good. And if you're looking
for tools like this, there are a lot of preset
prompts available inside Vizcom for things
like flash photos, putting it in a concrete room
or creating a mood board. Make sure to have a look at
the presets Vizcom offers as they may solve some issues you're trying to figure
out on your own.
19. Projects: Now that we've been
through all the content, I wanted to give you the
opportunity to try out everything you've learned
with some specific examples. On the screen, you can
see three objects, a set of headphones, water bottle, and a more
miscellaneous rectangular shape. Based on everything that
we've been through, have a go at putting all
three of these images through the process of trying to develop them into a fully
rounded product idea. It might be for the headphones, for example, you end up
removing the cables, changing parts of the headband, or completely rethinking how it actually goes on your head. For the water bottle,
it may be more of an exploration of
color and material. For the third one,
I'm leaving it open to you to work out what objects you actually want to make this. It could be an alarm clock, it could be a camera
or a door sensor. There's really no limit
to what it could be. So have a go at
putting it through the process and seeing
what you can come up with. Make sure to upload
all of these to the learning area so we all get to see what
you've created.
Theo Tucker, Designer - CGI Artist
