01

low poly modeling

Character Production Chapter 01 – Low Poly Modeling
Chapter 01 – Low Poly
Modeling
Software used: Maya
Basic Modeling
Hello my name is Anto Juricic and I am a CG
artist specializing in the creation of believable
CG characters. In this series of tutorials I am
going to walk you through the numerous tasks
involved in creating a highly realistic render of
an old man. As a core application I have chosen
Autodesk’s Maya and I will use it for the majority
of tasks in this tutorial.
Although Maya is a powerful tool by itself and
capable of both geometry sculpting and texture
painting, it is capable of creating better results
when used in conjunction with applications like
Photoshop and ZBrush, which are going to be
our major texture creation tools. With that said
let’s move on to the first step.
www.3dtotal.com
Setting up Image Planes
The edge extrusion method is possibly the
easiest way to block out the basic shape of a
human head, and for that we will rely on a pair
of reference images that we are going to load as
image planes. I have provided you with a side
and front image of this character as a download
with this tutorial so you can use the images to
follow the tutorial (Fig.01).
You can use any reference image you like as
long as they come from the same person and
show a front and side view. Let’s first spend a
few minutes preparing the images so they align
perfectly once we load them into the Maya. Load
the front image into Photoshop and pull down a
few guides to mark the position of the eyebrows,
center of the eye, nose tip and bottom, upper lip,
center of the mouth, bottom lip, tip of the chin
and bottom of the chin. Also place a guideline
along the center of the face.
page 5
Now, using the Crop tool and its transform
controls, extend the image size to the right so
you can fit a side image in it. Place the side
image to the right of the front image and scale,
rotate and move it until you align all the facial
features with the guidelines matching the front
image. Don’t worry if some parts don’t line up
100% perfectly; it’s probable because of the lens
distortion of images.
If your model’s head is slightly tilted to one side
you can use the front on view to adjust and
correct the image. With this done remove the
guidelines and crop the images again, removing
the space around the images. Also create two
separate files, one for the front and one for the
side view, and save the files.
Now go into Maya. In the menu select a new
image plane and press Load Image. Select your
previously prepared front image. Repeat the
Chapter 01

same process with the side view. Image planes,
by default, are created at the origin of the
scene and crossing each other. To get image
planes away from the center and make some
space you will need to adjust the attributes for
the image plane. Select the front image plane
from the perspective view, go to the Attribute
Editor and pick the Image Plane1 tab. Search
for the Placement Extras options and center
the parameter. Set the third number box, which
represents the Z position, to 40. Repeat the
same process for the side image plane and type
in 40 in the first box for the X axis offset.
Poly Modeling
Now we have our preparation done it’s time
to lay out some polygons, so let’s start with a
temporary eyeball. You can find all your poly
editing tools in the Edit Mesh menu, and you
can add those you use most to your custom
shelf by selecting a tool while holding the Shift
and Ctrl keys.
Create a polygonal sphere and adjust it to match
the position of the eyes from your image planes
(both front and side). From the front view create
a cylinder with 12 sides, which is slightly bigger
than the eyeball, then place it in front of the eye.
We will use this cylinder to shape the first row
of polygons for the eyelids. Delete the polygons
www.3dtotal.com
Chapter 01 – Low Poly Modeling Character Production
responsible for thickness so you are left with
just a plane circle. Select the boundary loop
from this circle and extrude it outwards. Delete
the inside so you have a ring shape. Also make
sure that you have your normal’s facing the
right direction. Now roughly align the vertices,
following the front and side reference images, to
trace out the shape of the eye. These first few
steps are illustrated in Fig.02.
Now select the outer border edge once again
and repeat the extruding process, shaping it
roughly to mark out the shape of an eye socket.
Make sure you align the newly created geometry
from the side view as well as the front view.
Select the two edges that separate the upper
and lower eye lid, and extrude them towards
page 6
the center of the nose bridge. Align this from
the side view too. Now select the border edges,
except ones at the center, and extrude them
once again to circle the cheek and eyebrow.
This is a good time to duplicate this piece of
geometry to the other side of the face. Select
the geometry and go to Edit > Duplicate Special
and in the Scale option type - 1.000 to X axis
only.
Move the duplicate to the other side of the face.
Cut the bridge of the nose vertically by inserting
an edge loop. Don’t forget to always realign
the new geometry from the side view. Now let’s
block out a nose shape by extruding down the
nose bridge to the base of the nose (Fig.03).
Chapter 01
Character Production Chapter 01 – Low Poly Modeling
Make seven extrusions along the nose and then
one more for the nostril. Select the second,
third and forth edge from the nose’s bridge and
extrude them towards the edge of the nostril.
Insert three edge loops vertically to the new
geometry and adjust it using the front and side
view as a reference. Don’t be afraid to go to
the perspective view from time to time to check
your progress in 3D and maybe make some
adjustments. Select the last edge from the
base of the nose and last edge from the nostril,
and bridge them to close the shape. Select six
edges starting from base of the nose all the way
around the nostril and extrude them out. Merge
the last vertex with the corresponding vertex on
the nostril, as illustrated in Fig.04.

Split the edge at the center of the long polygon

that bridges the nostril and nose base, and
reshape it to a circular opening. Bridge the
polygons to fill the gap between the nose tip and
the mask shape.

Select the nostrils’ opening edges, extrude

them a few times and translate them upwards to
shape the inside of the nose. With this complete
let’s move on to the mouth.
Extrude two lines from the base of the nose to
the mouth, then make ten extrusions circling the
mouth. Split the newly created ring by inserting
a edge loop at the center. Add one more loop
to define the mouth line. To finish the basic
mouth shape, extrude the inside loop to form
the mouth’s thickness. Split the long polygon
that bridges the mouth and nose by adding two
edges, and extrude one more circle around the
mouth. Now extrude two edges at the corner
of the mouth toward the cheek. Split these two
faces twice near the mouth’s corner. Align these
edges to the nasolabial fold using the front and
side references. Bridge the two faces from the
laugh line to the nostril, as illustrated in the last
step of Fig.05.
From the base of the mouth make five
extrusions to outline the chin, and bridge the
laugh line with chin, as illustrated in Fig.06.

www.3dtotal.com page 7 Chapter 01


Cut along the polygon three times and fill in the
gap. Add two edges to cut along the laugh line
polygon, and fill the gap by bridging the mouth
and the base of the nose.
Bridge the two polygons from the nostril to
the corresponding polygons on the eye mask.
Extrude two edges from the side of the mask
to create a temporal area. Do the same for the
cheek and add a few extrusions from the cheek
to the neck. Add one more edge to the cheek
polys and connect it to the last polygon from the
www.3dtotal.com
Chapter 01 – Low Poly Modeling Character Production
chin extrusion. Select five edges from the top
of the mask and extrude them twice to form a
forehead shape (Fig.07).
Split the long temporal poly and bridge it to
the forehead. Continue extruding the forehead
all around the head, all the way to the neck.
Reshape the new geometry to form a smooth
shape. Insert another edge loop to the cheek
geometry right behind the laugh line and
connect the cheek to the eye geometry. Do the
same to connect the cheek and chin. Don’t be
page 8
afraid to reposition some parts of the geometry
at any time, and add or remove some of it if you
feel it’s necessary. Make additional splits to the
cheek geometry to match the edges of the jaw
geometry. Fill the gap between the cheek and
jaw by bridging between faces. Use the same
technique to fill in the rest of the gaps, as shown
in Fig.08.
Also connect the remaining opened faces,
as illustrated in Fig.09. Select an open face
from the back of the neck and make enough
Chapter 01
Character Production Chapter 01 – Low Poly Modeling

extrusion to match the jaw line and fill the gap.

Select the open loop on the neck and extrude
down as many times as necessary to ensure
equal distribution of geometry. Also make sure
you delete the history from time to time.

Now go to vertex mode and press B on the

keyboard to adjust the falloff radius, which is
going to help us do some proportional editing.
Go to perspective view and have some fun with
this tool until you are happy with your model. I
hope you have enjoyed this tutorial; see you in
the next chapter where we will be adding some
detailed facial features. Check out Fig.10 for my
result after some proportional editing.

Anto Juricic
Web: http://anto-toni.cgsociety.org/gallery/
Email: monty.band@gmail.com

www.3dtotal.com page 9 Chapter 01

 02
modeling the features
Character Production Chapter 02 – Modeling the Features
Chapter 02 – Modeling
the Features
Software used: Maya

Hello! In the last chapter we used some very

simple techniques to block out the main form
of our character and establish a well organized
edge loop structure. The method I demonstrated
in the previous chapter will enable us to
continue adding more geometry, which will add
some details to specific facial features.

Perhaps the most important part of a face are

the eyes and the area around them. We are
so used to seeing people’s eyes that making
believable CG eyes is not an easy task. So let’s
continue with that area.

The eyeball itself will consist of two parts: the

inner one with a slightly concave iris area and
a slightly bigger outer one with a bulge over the
iris beneath it.

As a starting point you can use the sphere that

we created at the beginning of the first chapter
or create a new one from the front view. Make it
slightly bigger than the eye socket opening and
try to center it using a front image plane.
A default poly sphere comes with 20 triangles
at the pole, which makes an unwanted pinch
on the geometry. To get rid of these problems,
select every other edge radiating from the pole
and delete them.
Now instead of 20 triangles you have 10 quads
that will render much more smoothly.
Select those ten faces and grow the selection
twice, use the Scale Transform tool to scale
down selections on the Z axis to make it slightly
concave. Move it back into place on the Z axis.
Add one more edge loop at the edge of the iris
to straighten the edge.
To make the outer eye geometry, duplicate the
one you just created and scale it up just enough
to cover the inner part. Select all concave
polygons and this time scale them up on the Z
axis to make a bulge. Also use Move Transform
to get the bulge back into place.
Select the inner area and add the outer
geometry to the selection, then press P on the
keyboard to make a parent connection. The last
few steps are illustrated in Fig.01.
Let’s continue by shaping the eyelids and area
around the eye. When we have the eyeball in
place we need to reposition the geometry of the
eye opening to conform to a spherical eye. Don’t
expect to make it a perfect fit at this point since
there is more geometry to be added. Also, if
necessary, rescale the eyeball. It is a good habit
to arrange objects in separate display layers so
you can easily hide and unhide any object at
any given time. Hide the eyeball to make some
room for the next few steps.
Select the border loop of the eye opening and
extrude the selection inwards to make the eyelid
thickness. Scale down the last selection slightly
to make an easier angle transition between the
eyelid shelf and the rest of the geometry.
Next extrude the opening edges once more
and scale outwards to make geometry that
will penetrate the eyeball and prevent us from
having any gaps between the head and the
eyeball.
Select and delete three faces at the inner corner
of the eyelid to make a place for the tear duct.
Bridge the open edges left over from the
deletion and, using the Cut tool, cut in a shape
as demonstrated in Fig.02.

www.3dtotal.com page 11 Chapter 02

 Chapter 02 – Modeling the Features Character Production

Now it’s a good time to add some more

geometry to both rings and the loop surrounding
the eye.

Let’s start with cutting the rows radiating from

the eye. To add a complete edge loop all the
way, use the Insert Edge Loop tool. Newly
created edges are marked in Fig.03.

New edges are added linearly and need to

be repositioned to smooth out the surface.
To evade tedious vertex pulling and pushing,
the best strategy would be to use the Sculpt
Geometry tool.

Use the Pull brush with Auto Smooth checked

and the Smooth Strength set to 3. Set the
intensity down and swipe gently over a new
geometry.

Now we should add a few loops circling the

eye. Added loops are also illustrated in Fig.03.
Repeat the same geometry sculpting process
to smooth this area. Add more loops using
the same technique to even out the overall
geometry density.
Now things are getting more complicated and
at this point you should have more than enough
geometry to shape out all the major features of
the face.
At this point I am using the Move tool with Soft
Selection and the Sculpt Geometry tool to
shape out most of the facial features and fix the
proportions. Patience is your best route when
it comes to this task and it can often take a fair
amount of time to reach proportions that you’re
satisfied with. Don’t forget that you can always
come back to the proportioning at any time, so
it’s not a big deal if you don’t get it right the first
time.
Later in the process, if you think your mesh
is too dense, feel free to optimize it as you
like, but first make sure every edge is serving
its purpose, and contributes to the form and
construction of the mesh.
Fig.04 illustrates my result on the same mesh
after pulling and pushing the geometry, and
some mesh sculpting inside Maya.
Before we continue with more detailing it’s
important to have all of the head features
complete.
Once we attach the ear to the head model we
will become completely ready for final tweaking
and proportioning. So let’s continue with the ear.
You can hide all objects at this time and leave
only the image planes visible. You can use the
same image plane as you used for the head, or
if you like you can make a new one with an ear
image of your own choice.
Modeling the ear is all about laying out correct
topology that will follow the major ear shapes.
The ear is a complex shape and to better
understand the shape I strongly recommend you
learn the Latin names of the ear features.
Use the Create Polygon tool from the Mesh
menu to create a single quad polygon, place
it at the start of the ear lobe and extrude it all
along the helix, as illustrated in Fig.05. Repeat
the same process for the inner part, which is
called the antihelix. Do your best to match the

www.3dtotal.com page 12 Chapter 02

Character Production Chapter 02 – Modeling the Features
number of polygons to the helix and the edge
placement. Try not to add too many polygons at
the beginning; keep it as simple as possible.
Select all the polygons and extrude them
twice; on the second time extrude with a
slight offset. Now delete the polygons marked
with red arrows from both shapes and all the
polygons from the back, and then bridge the
corresponding edges. Also add one more edge
loop to slice the inner shape. For the last few
steps, check Fig.05.
Select the loop in the ear hole and extrude it a
few times to shape that area. Using the same
technique, extrude the outer edge border to
shape the back of the ear.
Now use the same strategy as we did for the
face, reshape the existing geometry by using
Soft Selection with the Move tool and Sculpt
Geometry tools (Fig.06 – 07).
Now unhide the head geometry, select the
edge loop on the opening and extrude it twice
to prepare the head for connecting with the ear.
Select Average Vertices from the Mesh menu to
smooth out the newly created shape. Place the
ear in position and start to use Soft Selection
with Move Transform to reshape the opening on
the head to fit the ear better. Combine the two
objects into one mesh.
www.3dtotal.com
You will find that the two objects do not match
in terms of topology and edge placement, and
you might reduce some of the edges from the
back of the head or add some to the ear before
merging an ear to the head. Start from the part
facing towards the front by selecting close-
matching edges and bridging between them.
This way all non-matching edges will stay at the
back of the ear, which is unlikely to be visible.
Do your best to keep the polygons four-sided.
After some tweaking, I’ve got the result
illustrated in Fig.08.
After successfully attaching the ear to the head
let’s move on to finalizing the rest of the mesh.
Grab the edge loop at the border of the mouth,
opening and extrude it as many times as
necessary to create a mouth bag. At this point
duplicate the eye mesh to the other side of
the head and, if you like, make a few more
extrusions to the base of the neck. You will most
certainly find more edge loops than you need so
feel free to optimize the mesh as you like. You
can use the mesh I provided with this tutorial to
compare your result with mine. Lighter geometry
will be much easier to unwrap and prepare for
texturing, which is going to be our next step. Until the next tutorial, happy verts and polys!
When you have all your geometry in place, once Anto Juricic
again use the Sculpt Geometry tool and Soft Web: http://anto-toni.cgsociety.org/gallery/
Selection to evaluate the shape and proportions. Email: monty.band@gmail.com
page 13 Chapter 02
 03
unwrapping
Character Production Chapter 03 – Unwrapping
Chapter 03 – Unwrapping
Software used: Maya
In the last two tutorials we have covered the
creation of a head model, using some simple
modeling techniques in Maya. So the next step
is to prepare the model for texturing.
The most common way is UV mapping, so
before we start I will try to describe what UV
mapping is. UV mapping or unwrapping is a way
of representing a model in 2D space, so it can
be used as coordinates that will correspond to
the 3D version of the model.
In other words it’s a model unfolded and
flattened onto 2D space. The simplest way to
understand this concept is to imagine a simple
paper box you want to unfold and flatten onto
a desktop (Fig.01). The same concept can be
applied to geometry of almost any complexity.
www.3dtotal.com
There are a few rules that need to be followed
during this process. To unfold any model, we
have to make cuts, or rather seams, in order to
produce less distortion on the flattened model.
More seams can often lead to less distortion,
but the downside to that approach is that those
page 15
seams will show as discontinued texture edges,
so the best approach is a good balance of both.
It’s also important to say that UV space is 1:1
square ratio and in order to maximize texture
space, it’s important to keep UV chunks or
islands packed tightly and fitted in that range.
Standard UV space ranges from 0-1 in a U
direction and 0-1 in a V direction, and it’s often
referred to as 0-1 space (Fig.02).
Every UV piece has to have its own space;
no overlapping is allowed, except two or more
pieces that are identical and destined to use
same texture.
In that space, U refers to horizontal coordinates
and V refers to vertical coordinates. In the same
way 3D is described by X, Y and Z.
With that being said, let’s move on to
unwrapping the head we previously
modeled. The fact that we have built a model
symmetrically will pay off through every step of
creation, especially the process of laying out
UVs. Therefore we don’t need a symmetrical
side of the head, and the first step is to delete it.
Once we have completed the UVs for one half of
the head, we will mirror it over to the other side.
Chapter 03
 Chapter 03 – Unwrapping Character Production

Along with geometry, UVs will be copied too and

all that’s left is to mirror the UVs and to stitch the
two UV shells together.

The second part of the plan is to separate some

parts in UV space, like the ears and the mouth
bag and unfold them separately. Keep in mind
that models in UV space can be in as many
pieces as necessary.

Let’s start with the ear. Select the polygons of

the ear (Fig.03). The selected polygons are
mostly facing planar to the side view and we will
use planar projection as a starting point for this
part.

Go to the Create UVs menu and select Planar

mapping to make an initial projection. To
preview how your texture will be applied to these
coordinates, let’s apply a checker texture to this
part.

Make a simple blinn material and connect

checker texture to color. Also, in a 2D placement
node, make it repeat five times. You can select
unwrapped faces directly from the UV Editor and
apply the material to the selection.

As you can see, the squares on the checker are

a bit stretched because initial planar projection
has made the projection plane perfectly square.
To fix this, select UVs from the UV Editor and
rescale it by simultaneously watching over a
texture in a 3D view, until you make the checker
texture on the ear almost perfectly square.
Planar projection has done a pretty good job,
except for a few areas that are overlapping or
facing away from the projection, so the texture is
stretched. So let’s deal with those issues.
First let’s detach the ear canal and move it out
of the way. In the UV Editor, select one edge
inside the ear canal and from the Select menu
press Convert Selection to UV Edge Loop. From
now on the Select menu will be used frequently,
so detach it to make it a floating menu.
To make the cut, go to the Polygons menu in
the UV Editor and choose the Cut UV Edges
option. The ear UVs are now separated into two
shells and to move the ear canal out of the way,
select one polygon from that shell and press
Select Shell, then Convert Selection to UVs.
Now you can move the separated piece to the
side (Fig.04).
With the ear canal shell selected, go to the
Tools menu and then select the Smooth UV
tool. From there you will be presented with two
on-screen menus, one called Unfold and the
other called Relax. Both of these are smoothing
options. My strategy is to apply Unfold, then
Relax, in small increments until I get rid of all the
overlapping.
You can also move vertices in UV space
one by one and help those two algorithms to
better understand what you are up to. After
unwrapping that part, rescale the shell to
approximately the same texture space as the
ear shell. In order to do so correctly, use 3D
View to match the checker size on both shells.
Now let’s concentrate on the ear, which is the
much more complicated part. In order to use the
Unfold and Relax technique correctly, we need
to make some unfolding by simply moving some
vertices into the UV Editor. So let’s start with
that parts that are most obviously overlapping,
such as the outer edge of the shell. To select
the shell border, select one of the outer vertices
and press Select Shell Border.

www.3dtotal.com page 16 Chapter 03

Character Production Chapter 03 – Unwrapping
Deselect the inner border, since we will
concentrate on the outer edges of the shell first.
To move selected points out of the way, we will
use the UV Smudge tool, which is essentially a
move brush with a nice falloff. Once you move
these points to their own UV space, grow the
selection once (Shift >) and repeat the same
process again.

Now let’s deal with the rest of the overlapping

vertices by selecting them in small groups and
relaxing. Make sure to constantly monitor the
behavior of the checker projection in 3D view, to
prevent any major texture distortions (Fig.05).

Since the ear is complicated in both topology

and shape, making perfect distortion free
coordinates is close to impossible and texture
stretching is tolerable on this part, since most
of its look in final rendering will come from its
shape and sub surface scattering shading.
Once unwrapped, select both parts and move
them out of the way to make room for the
next piece. Now select all the polygons of the
mouth’s interior and make sure that no polygons
are selected that are visible from the outside.
That way we will prevent any visible seams.
Apply planar projection also on this part as a
starting point and apply a checker material also.
Select UV Shell and apply the Unfold option all
the way to the maximum. Use the UV Smudge
tool to deal with any overlapping remaining
(Fig.06).
Repeat the same process for the geometry
inside the nose and move the unwrapped parts
out of the way. Now we are left with the main
part of the model. To easily select the remaining
part, go to the UV Editor and select all the
unwrapped pieces, then go to Edit and choose
Invert Selection. With this area selected, go to
the Create UVs menu and select Cylindrical
Mapping.
Since the shape of the head closely resembles
a cylindrical shape, this is the best starting
projection we can get.

Here comes a tricky part. Because we have

deleted half of the head it’s no longer fully
cylindrical, but half of it. This issue can be easily
fixed by rotating and moving projection in 3D
view by using a projection gizmo (Fig.07).

Here is where the checker texture will be most

valuable because we will constantly refer to
it in a 3D view, while correcting distortion and
overlapping in the UV Editor.

Apply the checker texture to the whole model

and resize the checker size by setting up more
repetitions in a 2D texture placement node.
Between 30 and 50 repetitions would be a
good number of repetitions. Resize all the other
unwrapped parts to roughly match the checker
size with the head.

www.3dtotal.com page 17 Chapter 03


If you have done everything right by now your
checker projection should look like this (Fig.08).
As you can see there is some obvious stretching
of texture in some places, so let’s deal with
those major issues first.
Stretching in general occurs at places where
the size of the polygons in the UV Editor doesn’t
match the size of the same polygons in 3D view.
Our job is to make those differences as invisible
and as minor as possible.
Let’s correct the most obvious part at the
neck. Use the UV Smudge tool to space out
overlapping vertices as much as possible, then
select points in small groups of up to 10, and
apply Unfold and Relax smoothing alternately.
Repeat the process for all of the neck area until
you remove most of the distortion. After that
select all points in the neck area, except ones at
the border, and repeat unwrapping and relaxing.
Similar to the neck area, the top of the head
will also have significant distortion, but since it’s
not going to be visible at all, don’t spend much
www.3dtotal.com
Chapter 03 – Unwrapping Character Production
time on it. Just select the top vertices and apply
Unfold smoothing.
The next problematic area is the mouth, where
we have some overlapping. The best approach
to this area is to move the points one by one,
since there are few of them.
Here is what I came up with after applying the
last few steps (Fig.09).
Using a similar approach, relax the areas on the
nose and eyes too. Detach the inner part of the
eyelid, as it’s not going to be visible and it will be
easier to unwrap the eye without that part. Try to
maintain the shape of the eye in the UV Editor;
it will be much easier to draw in 2D when you
have them similarly shaped in both 2D and 3D.
After completing UVs for half of the head, it’s
time to mirror all this work to the other side. Go to the Mesh menu and select options for
Before mirroring, align all the points at the mirroring the geometry. Make sure that the
middle of the head in the UV Editor, so they are Merge with the Original option is unchecked and
aligned vertically. Apply.
page 18 Chapter 03
Character Production Chapter 03 – Unwrapping
Select the vertices at the middle of the head and
go to the Edit Mesh menu. From there select
the Merge Vertices option and set Threshold to
0.0010 and hit Apply.

Now you have the whole head unwrapped and

both of the sides are using the same UV space,
which means that the texture applied to the one
side will also be applied on the other side, but
it will be mirrored. That can be good in some
situations, like in game models, but for this
purpose we will make both of the sides use their
unique UV space (Fig.10).

Using a 3D view, select one polygon from the

mirrored side and in the UV Editor go to Select
Shell. With the mirrored shell selected, choose
Flip from the Polygons menu in the UV Editor
and finally, from the Select menu, use Convert
Selection To UVs. Now you have selected only
UVs for the mirrored side and you can move that
shell left to match the other side.
Select edges at the center of the head in the
UV Editor, and from the Polygons menu use
Move and Saw UV Edges to stitch the two
parts together (Fig.11). Also make sure to flip
the shells for the ears, mouth interior and rest
of the pieces. Stitching both sides together will
produce some stretching at the line where both
shells for the head are welded.
Before packing all the parts into 0-1 space,
invest some more time in relaxing those areas.
The last step is to arrange all the shells to fit in
0-1 space nice and tightly.
I hope you enjoyed this tutorial and I’ll see you
next time.
Anto Juricic
Web: http://anto-toni.cgsociety.org/gallery/
Email: monty.band@gmail.com

www.3dtotal.com page 19 Chapter 03

 04
texturing and shaders
Character Production Chapter 04 – Texturing and Shaders
Chapter 04 – Texturing
and Shaders
Software used: Maya, ZBrush
In the last tutorial we made all the necessary
preparations for texturing, by unwrapping the
model so that we have clean and readable UV
coordinates. That way we can be sure all the
details we draw in to texture will wrap around
the model properly.
The best way to start any texturing project is to
conduct a good search for reference images.
You can start with the well-known www.3d.sk
reference site or just browse Google for some
faces that you like. Sites like www.facity.com
can provide you with hundreds of useful face
images from all around the world. To use these
references easily, compose a collage of your
favorite images in one file.
There are many ways to achieve believable
surface details, but the most intuitive way for
me is by sculpting surface details in ZBrush.
Although we have finished modeling, shaping
and even UV mapping, it’s never too late to
review your model once again, so feel free to do
so if you think it needs any improvements.
I could share with you some standard ZBrush
brushes I use on a daily basis, but let’s see how
you can make one all by yourself. In the next
few steps I will describe how you can make a
realistic skin pore brush, so let’s jump to ZBrush.
www.3dtotal.com
Make a new ZBrush document with a size of
1024 x 1024.
In ZBrush 3D meshes you will find a simple
polygonal plain. Drop it on to the canvas and
press Make Polymesh 3D, then subdivide the
plane six or seven times to ensure they’ll be
enough mesh density to receive pore details.
From the Brush palette choose Standard brush.
Change the stroke to spray, and pick alpha 47.
Frame the model to fill in the canvas. You can
frame in any spot on the plane or the entire
plane, it’s up to you. Press Alt and swipe the
whole visible area of the plane with this brush to
make it look really noisy (Fig.01).
Now go to the ZBrush Lightbox menu and from
the brushes folder, scroll right to the Smooth
brushes where you will find the Smooth Peaks
brush. Use the Smooth Peaks brush and swipe
all over the noisy area once again. The surface
that you achieve this way should look a lot like
orange peel or cheek skin.
In the Alpha menu, choose Grab Doc and save
the captured alpha as a PSD file. The saved
alpha image stores depth information and once
loaded to the Standard brush with a rectangular
stroke, it will be valuable in detailing the skin.
Before applying all these high frequency details,
let’s concentrate on some larger forms, like
page 21
wrinkles and folds. Now would be a good time to
study the references you gathered earlier.
Loading images in Spotlight is the easiest way
to use references side by side with the model.
All you need to do is load the image using the
Texture menu and press Add To Spotlight.
Spotlight is very easy to use and all the options
will be self-explanatory. Lastly, go to the brush
settings and under the Samples tab, turn off
Spotlight Projection.
Now it’s time to load the object file exported
from Maya. Go to the Tools menu and import it,
then drop it onto the canvas and press T.
In order to start adding details, we have to
subdivide the model a few times to get a denser
mesh. The dividing process tends to smooth
the mesh significantly and to prevent losing any
mesh volume you must store a morph target
before adding divisions. After achieving the
wanted level of divisions, press Switch in the
morph target tab to regain the original volume.
The next task is to emphasize the shapes you
already modeled in the mesh and add new
shapes like skin fold and wrinkles. I like to use
the ClayBuildup brush with alpha 48, and for
fine wrinkles I use the same brush with the
LazyMouse option.
The best advice I can give you at this stage
is to be patient and don’t rush to the highest
Chapter 04
 Chapter 04 – Texturing and Shaders Character Production

subdivision level, as long as your current level

has enough resolution to describe your strokes.
Also keep your eyes on the references. You
don’t have to invent new skin folds or wrinkle
patterns – nature has done it for you. All you
have to do is to copy what you see. Observation
is the key.

Fig.02 illustrates my result after half an hour of

doodling with the ClayBuildup brush on the third
subdivision.

As you can see there are not any major changes

to the volume of the model and all added details
are a thin layer on the surface. The wrinkles
you add at this stage will guide you through
the rest of the process, so step up to the next
subdivision and continue defining those details.

Don’t neglect any part of your model and make

sure all your details are at the same stage of
development on all areas of the model. You
should have enough resolution at this time to
start cutting in wrinkles and the best brush for
that task is the Dam_Standard brush, an brush
that pushes and pinches geometry at the same
time. Fig.03 illustrates detail cuts achieved with
the Dam_Standard brush.
All of these details so far are on the fourth
subdivision and since I feel like the mesh
resolution does not support any finer details, I
step up one division. The subdividing mesh will
most certainly wash out some of the lines, so
you will have to go over all the mesh and define
those details once again, by using the same
techniques as before.
As you can see by now, sculpting details is all
about layering through subdivision and the best
advice I can give you is to stay cool and don’t
rush for the multimillion polygon count too soon.
Now it’s time to use that skin pores alpha we
prepared at the beginning of this tutorial and
set up a pore brush. Select the Standard brush
from the Brush palette and change the Stroke
to DragRect. Go to the Alpha palette and import

www.3dtotal.com page 22 Chapter 04

Character Production Chapter 04 – Texturing and Shaders
the earlier prepared alpha. In the Alpha Modify
menu change RF (Radial Fade) from 0 to 15
and the pore brush is ready. As a last step,
change the Smooth brush to Smooth Peaks.

You can start adding pores at the fifth

subdivision level, with subtle Z Intensity (about
10) and cover your entire model. Dragging the
pore brush on the surface will give you direct
control over the location and size of pores. You
can also use it with Alt pressed to achieve the
opposite effect. If you feel like you’ve overdone
it on some areas, use the Smooth Peaks brush.

Fig.04 shows my result on the fifth subdivision.

For the final detail, divide the geometry once
more to get a final, sixth level of subdivision
with approx 3-4 million polygons. You can go for
an even higher poly count if you have enough
RAM, but since the final goal is to project all
these details on 4096 x 4096 Displacement and
Normal map textures, I am sure those extra
details will not show up on a 4K texture. add more variety by adding individual pores you can come back, adjust the layer opacity
in some places. It’s a good idea to keep these and generate a new Displacement or Normal
Proceed with adding skin pores on level six, details on a layer in ZBrush, so if you at any map. Keep in mind that once rendered with the
using same technique as before. You can even point feel like your surface is looking too rough, subsurface scattering shader, most of the high
frequency details will get washed out, so don’t
be afraid to make those details a bit stronger
than you perceive them, by looking at some
photos of faces.

Fig.05 illustrates the final detail level on the

sixth subdivision.

Finally it’s time to transfer all of this hard work

into a Displacement and Normal map. We will
use both maps as a guide for texture painting in
Photoshop and we will also extract some details
out of both maps and use them in a mixture with
other color layers.

To generate a Displacement map, first you have

to go to the first subdivision level. The next step
is to choose the image size under the UV Map
tab – the 4096 preset should be enough for this
tutorial.

www.3dtotal.com page 23 Chapter 04


Go to the Displacement map tab and apply
the following settings: Adaptive, Smooth UV,
and DPSubPix 2. Press Create Displacement
and once it’s created press Clone Disp and
save it from the Alpha menu as a TIF file. The
generated map will be vertically flipped and you
can flip it right away from the Alpha menu or in
Photoshop.
Go to the Normal map tab and create normals
with the following settings: Tangent, Adaptive,
Smooth UV and SNormals. Export the Normal
map in a similar way as the Displacement.
Generating maps with these settings can take
some time to render, so it can be a great time to
take a break.
Before you export the geometry too, you can
use the Move brush to break up the symmetry
or make some facial expressions.
Now let’s move to Photoshop to draw a Color
map. Open a new document, with a size of 4096
x 4096 pixels. I usually start with a big chunk of
skin taken from a photographic reference. On
top of that, add the Displacement map. With
the displacement layer selected go to the Filter
www.3dtotal.com
Chapter 04 – Texturing and Shaders Character Production
menu > Other > High Pass. Set the High Pass
radius to 10 pixels.
Change the blending mode for the displacement
layer to Linear Light. Now we have darkened
cavities and have also laid down the road map
for the rest of the texture. To make this layer
blend more naturally with the skin, use Curves
adjustment and change the parameters on the
red channel only, as illustrated in Fig.06.
Let’s add some fine details we are going to
extract from the Normal map. Open the Normal
map as a separate file, go to Channels and
select the green channel only. Press Ctrl + A to
select all the pixels from that channel and copy
them. Now go to the texture file and paste to a
new layer. Let’s repeat the same process with
the High Pass, but this time set the radius to 2
pixels.
Overlay this layer in Soft Light blending mode.
This will be a base for a color texture, so before
moving on to other layers, group these layers
together.
Here is a quick setup method for making a
spray color brush for adding skin variation and
blemishes. Press F5 to get to the Brush Setup
page 24
menu and under Brush Tip Shape, choose a
5 pixels dot and adjust Spacing to 50%. Next
check the Shape Dynamics and set the following
parameters: Size Jitter to 100%, Minimum
Diameter to 10%, Angle and Roundness Jitter to
100% and Minimum Roundness to 50%.
Now move to the Scatter options and set Scatter
to 1000%, Count to 2 and Count Jitter to 50%.
Once again it’s time to pull out some of those
gathered references and start spraying some
variation over the skin.
I like to start with the red color and spray it all
over the place, and later concentrate one more
pass on the areas like the cheeks and nose.
Repeat the same process with a white color
on a separate layer. After adding some white
pigmentation, open a new layer and do the
same with a blue color, but more subtle and on
localized areas like under the eye area, nose
and cheeks.
After you finish all these color variations you can
flatten the noise layers and apply a Gaussian
Blur with 0,3 px radius to make it look more
softened. See Fig.07 for a small patch of the
skin texture achieved this way.
Chapter 04
Character Production Chapter 04 – Texturing and Shaders
At this point it’s a good idea to check the texture
on the model, so go to Maya, load the model
and assign a Blinn material to it. Load the Blinn
material with the PSD color texture and Normal
map to have a better representation of your
texture in Maya’s viewport. Fig.08 shows how
your model and texture should look in the Maya
viewport.
Keep on coloring the texture with more color
variations, skin blemishes and spots. Once you
are satisfied with the result, add a color balance
on top so you can control the overall skin tone.
Now we have all the necessary assets for
setting up realistic skin shading in Maya using
the V-Ray render and its powerful SS2 shader.
First we have to make some lighting, properly
set up the Displacement map, make a new
Maya scene, load a base model exported from
ZBrush and apply a V-Ray material. Let’s set up
a three-point lighting scheme by adding three
V-Ray lights: one key, one fill and one back
light. Use the V-Ray rectangular light and place
them as shown in Fig.09. To see an almost
instant effect of your lighting, use the V-Ray RT
engine.
To set up a displacement for the mesh, select
the object, go to the Attribute Editor and from
the V-Ray menu use the Subdivision option. The
first thing you will notice if you make a render at
this time is that the model renders smoothly.
www.3dtotal.com
Go to your Shading Group Attributes of the
V-Ray material applied to the model and
connect the displacement node to Displacement
Mat. Load the Displacement map you earlier
exported from ZBrush. In the File attributes of
the loaded map, set Filter Type to Off to avoid
any image processing and loss of details. Now
go to the Color Balance, set the Alpha Offset to
-0, 5 and check Alpha is Luminance.
Now you can make a test render. The reason I
am doing this with a V-Ray material instead of a
SS2 material is because subsurface scattering
tends to soften details, so it would be hard to
make the right decision on the displacement
strength. If you feel like you should have a
stronger displacement, amp the Alpha Gain
number and make sure that Alpha Offset is
always negative half of that number. See Fig.10
for my result after applying displacement.
Now, when we have a proper displacement
setup, let’s apply the Fast SSS2 shader to the
page 25
model and drop the same displacement in the
Displacement Mat. slot of this material. V-Ray
Fast SSS2 is really easy to set up and it has
already prepared presets for a few different
subsurface models like skin, milk, marble and
so on. For our purpose the skin preset will work
perfectly and the first thing you should do is set
the Prepass rate to 1 and load the Color map to
Sub-surface Color.
You can also plug in a grayscale map to control
the specular reflection and glossiness. You
can also use the V-Ray Texture Input Gamma
attribute. For an in-depth look at the material
parameters, refer to Fig.11.
That would be all for this tutorial and I hope you
have enjoyed it.
Anto Juricic
Web: http://anto-toni.cgsociety.org/gallery/
Email: monty.band@gmail.com
Chapter 04
 05
hair and clothing
Character Production Chapter 05 – Hair and Clothing
Chapter 05 – Hair and
Clothing
Software used: Maya, ZBrush
In this fifth tutorial, we will look at some poly
modeling techniques, very similar to ones we
have used to block out the basic head shape,
but this time we’ll model some cloth for our
character.
After that, to finally complete all the assets for
this work, we’ll add some hairs to the character’s
head using a few different techniques such as
Paint Effects and Maya Fur.
Let’s start with blocking out a basic shape for
the shirt.
As opposed to the edge extrusion method we
used on the head, now we’ll use box modeling.
Instead of gradually building up a volume box,
modeling usually starts with the volume. So
www.3dtotal.com
let’s start with a simple polygonal cube, roughly
shaped to the volume of the character’s torso,
as illustrated in Fig.01.
Split each face by adding a vertical and
horizontal edge loop to one of the faces, and
delete the top and bottom faces. Select the
open edge loop on top and extrude to form a
loop of faces around the neck. With this basic
shape blocked out it’s time to add some more
geometry. Go to the Mesh menu and choose the
Smooth option to divide the geometry once and
smooth the shape. After that use Soft Selection
page 27
to reshape the model until it looks similar to
what is shown in Fig.02.
If you like to use a more intuitive way for
reshaping, I strongly recommend that you
download the Diamant Modeling plugin by the
very talented Richard Scott Diamant. Besides a
handful of amazing features, you will find a grab
brush similar to one you can use in Mudbox.
Delete six faces next to the shoulder area to
prepare some room for adding a sleeve. Add
one more edge loop between the shoulder and
collar area. Select the sleeve’s opening loops
and make one extrusion. Use the Average
Vertices function from the Mesh menu to
smooth the newly created geometry. Again, use
Soft Select transform to reshape the geometry.
At this point you can duplicate this part of the
geometry and save it as a starting point mesh
for the vest, which we’ll concentrate on after
the shirt. The last few steps are illustrated in a
Fig.03.
Chapter 05
 Chapter 05 – Hair and Clothing Character Production

Add a loop around the collar and the middle of

the shirt, as illustrated in Fig.04, and extrude the
sleeve from the opening that we prepared for it
earlier. Mirror the geometry to the other side and
weld the vertices on the back of the shirt. Use
the Soft Selection transform to overlap one side
of the shirt over the other.

Next select all the faces and use the Extrude

function to add a thickness to the shirt. Then
delete all the inside faces that are not going to
be visible.

Now let’s prepare the geometry for the collar.

Select the loops on the faces surrounding the
neck area and duplicate the faces. See Fig.04
for a step-by-step illustration.

Select the edge loop from the upper side of

the new geometry and extrude once out on
the X and Z axis. Reposition the extruded
edges to roughly shape out the collar shape.
Add a couple more edge loops to even out the
geometry distribution and reshape the geometry
to the final collar shape.

When we have a basic shirt shape modeled, it

is time to add some detail to the cloth seams.
Before going into detailing, add edge loops
to the collar and shirt edges to strengthen
the edges, then add appropriate loops in the
shoulders to add seams, as shown in Fig.05.

Now when we have got all the geometry in place

it’s pretty easy to fine-tune the overall shape and
add more details.

After unwrapping this model I have separated

the sleeves from the shirt to get a more
pronounced seam. To make the shirt look more
natural, add a few more divisions to the mesh
with the Smooth function from the Mesh menu,
and use Maya’s sculpting tools to add a few
wrinkles. You can see my final shirt in Fig.06.

To start with the vest, let’s get back to the basic

shirt model and reshape it to roughly the shape

www.3dtotal.com page 28 Chapter 05

Character Production Chapter 05 – Hair and Clothing

of the vest with the V opening. Note the marked

polygons in Fig.07, which are important for
adding seams to the vest.

If you have successfully finished the shirt

part you won’t have any trouble with the vest,
because it’s pretty much the same in terms
of modeling techniques. After blocking out a
basic shape, add thickness to the geometry
and delete all the unnecessary faces. Add edge
loops at the seams and open areas to sharpen
up the details.

Don’t forget to unwrap the model before moving

on to the sculpting, where you will add some
looseness to the model to make it look more
natural, as I’ve done in Fig.08.

Finally, we reach the last piece of the clothes

– the cap. This part will be straightforward
because of its very simple shape. To start
with, make a simple box and apply the Smooth
function to add two more subdivisions. With this
method you will get a poly sphere and you need
to delete the bottom half. After that place the
newly created cap shape on the model’s head
and reshape it to fit properly. These few steps
are illustrated in Fig.09.

After adding a few loops and reshaping to a final

shape, I’ve come up with the result shown in
Fig.10.

www.3dtotal.com page 29 Chapter 05

 Chapter 05 – Hair and Clothing Character Production

In the second part of this tutorial, we will

concentrate on making facial hair using several
different techniques available in Maya and
V-Ray. Let’s start with Paint Effects, which we’ll
use to make the brows and hair. Paint Effects
is a very powerful system that will enable us, in
this case, to grow geometry instances on top of
the polygonal surface and much more.

We will start with the brows. You can choose

Paint Effects either from the Shelf menu or from
the Rendering sub-menu > Paint Effects. Select
your head mesh and from the Paint Effects
menu, choose Make Paintable. Now all strokes
that we apply to the mesh will be snapped on
the surface. Choose Get Brush from the Paint
Effects menu, and from the Hairs folder choose
EyeBrowBlack preset.

This preset will serve as a good starting point,

so make a stroke in the shape of the brow on
the surface to generate the eyebrow. As you can
see from this first stroke, the brow doesn’t look
like something that would fit the size and type of
our character so let’s do some tuning.

Go to the attributes of the created stroke called

eyeBrowBlack1 and let’s start with the Scale
attribute. I’ve found that a scale of 4, 5 works
fine for me as a starting point. Now move down
to the Tubes menu and under the Creation
options adjust Tubes Per Step to control the
number of hairs generated on the stroke. I have
set mine to 0, 7, which is quite a small number
of hairs, but it will get much denser and more
natural once I apply multiple strokes. The next
parameter is Segments, which controls the
number of segments along each hair. The more
segments you have, the more sensitive the hair
will get to displacement and deformations, which
we’ll address later.

The next one is the hair length. Default values

are looking fine to me in this case, but you can
play with those to get any desired results. This
is the same with the Tube Width 1 and 2, which
stands for the root and the tip width. Values of

www.3dtotal.com page 30 Chapter 05

Character Production Chapter 05 – Hair and Clothing
0.004 and 0.003 work fine for me. Now let’s play
with some displacement.
There are quite a few options there, but I would
like to keep it as simple as possible so I will
only use the bend parameter for the sake of
this tutorial. Feel free to experiment and maybe
you’ll get more interesting results.
I have set Bend to - 0.500 and Bend Bias to
-0.300. It would be a good idea to save this
preset at this time, so go to Presets in the
Attribute Editor and save the brush preset. To
apply this setting to every stroke you are going
to make, choose Get Setting from Selected
Stroke from the Paint Effects menu.
Now you are ready to make some brows, you
can delete that starting point stroke and apply
several new ones to make it look really natural.
You can also adjust all settings individually to
www.3dtotal.com
each stroke you make. You can use the same
brush to make some hairs in the nostrils and ear
hole. Every stroke has its own material and to
use V-Ray materials you have to convert those
strokes to polygons from the Modify > Convert
menu. Now you can apply different V-Ray
materials to each stroke. Later on we will deal
with the shading of hairs. You can see my result
in Fig.11.
We will use the same technique for the hair, but
this time we will use a brow preset as a starting
point. First change the brush with the parameter
to a higher value, because it will be much easier
to generate hair with the wide stroke. I set my
value to 0.500. After that move down to the
Creation menu and adjust the hair length. I am
going for a more bushy hair style, so I’ll set mine
to 1 for the max and 0.500 for the min length.
I’ll also add a few more Tubes per step; a value
of 5.000 works fine for me. The last thing I’ll
page 31
address is the Displacement menu, where I’ll
play with some noise and wiggle options. Note
that since the hair is now much longer, you will
also need to increase the number of segments.
Once you are satisfied with your settings, don’t
forget to save out the preset. You’re now ready
to lay down a couple of strokes to populate the
hair area, as I did in Fig.12. You can use the
same approach to generate hairs for the beard.
That would be all for this tutorial, I hope you
found it useful and enjoyed it. In the next and
final tutorial we will polish the whole scene and
try some different lighting techniques.
Anto Juricic
Web: http://anto-toni.cgsociety.org/gallery/
Email: monty.band@gmail.com
Chapter 05
 06
rendering and lighting
Character Production Chapter 06 – Rendering and Lighting
Chapter 06 – Rendering
and Lighting
Software used: Maya, ZBrush
In this sixth and final chapter of the series, we
will put together everything we’ve done so far
in a nice and realistic render. Unfortunately we
didn’t have room to cover every single step in
the process, but the things I chose to leave
out are more or less self-explanatory, like the
texturing of the clothes and modeling eyelashes.
Once you explore the downloadable content
from this tutorial any missing parts will fit in to
place. So without further a do, let’s move on to
the assembling of the final scene.
Instead of recreating the image you can see at
the beginning of every tutorial, this time I will use
the actual scene to describe the rendering and
lighting process.
The good thing is that the content that you
can download with this tutorial will give you an
identical result to my final image. You can use
it to follow this tutorial or as a scene preset for
your own character. You will find some slight
differences in the scene assets, but that will not
affect the course of this tutorial.
Over years of work, I’ve found that the strategy
that works best for me is to start with lighting a
gray-shaded model, because it’s much easier
to read shape without all different colors and
www.3dtotal.com
materials. Once I am satisfied with the result,
I will make it more interesting with different
materials and textures.
Load all the assets you made so far for this
scene and assign a new V-Ray material to
all objects. Leave the material on its default
settings and name it something like “test
material”. This would be a great opportunity for
you to use the V-Ray RT render, which will give
you almost instant feedback.
For the easiest setup and most realistic result
I’ve decided to use IBL (image based lighting).
V-Ray can utilize IBL in a couple of different
ways, but the easiest to set up is a dome light,
which is an infinite big light that surrounds
everything in your scene. The default dome light
is white in color, which is a great way for adding
ambient light to your renders. Instead of a clean
white light we will color the light with a HDR
image. The effect of using light in this way will
result in very realistic environment lighting.
From the Create menu choose Lights and
create a V_Ray_Dome_light. Under the light
shape attributes scroll down to Texture menu
and upload a HDR image. For this example I’m
using the “kitchen_probe” HDR image, which I
downloaded for free from www.pauldebevec.
com/Probes/. You will find there are many other
examples of so-called light probes that also
work well with this or any other scene, and I
page 33
strongly encourage you to try a few other HDRis
as well.
At the top of the light’s attributes you will find a
small box named Sample, which will show you
the approximation of the light’s color intensity
and you can use it for setting up the initial
intensity for this light.
Once you load such an image into the dome
light, Maya will automatically create an image
placement node, where you can make some
additional settings, such as rotating a light or
choosing a type of mapping.
A Rotation setting of 126.00 units and Intensity
of 0.800 works fine for me and you can see my
first render with this setup in Fig.01.
To gain more control of your lighting, you can
experiment further with the light probe dynamic
range by shifting settings like Alpha Offset and
Alpha Gain. In Fig.01 an Alpha Offset of 0,500
gave me the best result. Image-based lighting
does not have to stand all by itself; you can
always add some additional lights to better
frame your character.
In Fig.02 you can see another example of the
image-based lighting; this time I’ve used an
image probe called “campus probe”, which is an
image of an outdoor scene with a little bit of low
sunlight and a yellowish color tint on the side.
Chapter 06
 Chapter 06 – Rendering and Lighting Character Production

As you can see you can get some really

interesting results with image-based lighting and
change complete scene lighting with one file.
Since dome light emits from all sides around
your model, it almost looks like the scene has
been lit with global illumination techniques,
adding a great sense of realism to the render.

Another good way of lighting your scene is with

another type of light, such as V-Ray rect. lights.

For the next example I will construct a basic

studio lighting setup, arranged with four V-Ray
rectangular lights. I will have two key lights (one
for each side of the face, placed on the left and
right), one fill light in front to lighten up dark
areas and a rim light for the back to give the
character some outline for better framing.

You can see the placement and colors of the

lights illustrated in Fig.03.

Just by tweaking the intensity you can get a few

different and interesting results, as shown in
Fig.04. Another good idea would be to parent
all light to a locator placed at the center of the
scene. By rotating a locator on a Y axis you will
be able to rotate the whole light rig around the
model.

Now, when you have your favorite light setup

blocked, it’s time to turn our attention to
materials. Since we have already covered skin
material shading in one of the previous chapters,
let’s say a couple of words about shading the
clothes.

All of the cloth pieces have some sort of

repeating pattern and naturally those patterns
are placed from seam to seam. To achieve
that kind of look with a small repeating pattern,
it’s essential to model cloth with the topology
following those seams.
simple rules, it’s quite easy to do the surface You can see how a few simple seamless
Another important step in the process is to make details by repeating a weaving pattern across patterns repeated across the whole vest work in
cuts in the UV shells along the seams. Once you the whole mesh without actual additional texture Fig.05. The same process can also be applied
have constructed the clothes following those painting. to the cap.

www.3dtotal.com page 34 Chapter 06

Character Production Chapter 06 – Rendering and Lighting
Further cloth realism is achieved with the
addition of some surface hairs. To make this
process as simple as possible I will work with
Maya Fur, which comes with some presets that
can be used as a starting point. After making
a few changes to the selected preset, I render
hairs with the Maya software renderer on a
cloth model with a black surface shader. Maya
Fur renders extremely fast with the software
render, and after less than three minutes I have
a fur pass to composite over a beautiful render
(Fig.06).
You can also experiment with V-Ray fur, which
is also great and very easy to set up.
www.3dtotal.com
To set up the V-Ray fur, select the object you
wish to grow fur from and go to Create > V-Ray
and Add VrayFur to your selection. You will
instantly get feedback of the created fur in the
viewport and can quickly set up a desired look.
Note that V-Ray fur will inherit material from
its base surface. If you want to have V-Ray
fur with different materials you can copy the
object, assign it to the other material and turn
off the object’s visibility in the shape node
object display. Fig.07 illustrates the detailed cap
surface with V-Ray fur.
Another important part of achieving a realistic
CG model is the creation of the character’s
page 35
eyes. There are a couple of important things
worth mentioning when it comes to eye creation.
An eye model can be constructed out of many
parts if you are going for an anatomically correct
model, but the easiest way would be to use two
models, one for the outer transparent layer with
the cornea and another for the layer with the iris
and sclera. These separated models are shown
in Fig.08.
The inner part should be shaded with
subsurface scattering and the outer part with a
fully transparent material with the index of the
refraction set to 1.3 or more.
After assembling everything together, the last
thing to do is to set up a render resolution and
sampling.
Chapter 06

Note that more image resolution requires less
precise image sampling. V-Ray offers us three
types of image sampling: Fixed, Adaptive
Subdivision and Adaptive DMC. Image sampling
could be a whole tutorial in itself, so let’s skip
the complicated stuff and cut to the basics.
I often use the Adaptive DMC method because
it works best with the small details, like hairs or
skin pores, which are important for achieving
a realistic image. If you are using low samples
you will have probably lost many fine details you
sculpted in your model’s face, so you have to
use fixed or low adaptive subdivision; only for
testing or if you are rendering a big resolution.
If you are going for a 1080 px video render or
similar resolutions I would definitely recommend
DMC with minimum 1 and maximum 6 samples.
You can see a couple of frames captured from
the turntable video in Fig.09 – 10.
Threshold parameter also plays a big role in
image sampling and you will get more precision
and less noise with a lower threshold.
www.3dtotal.com
Chapter 06 – Rendering and Lighting Character Production
If you are rendering big resolution stills, like Also using the same color correction
3000 px or above, you can probably get away adjustments for both the character and the
with default V-Ray settings, but I would definitely background will help you to marry the two
recommend using a DMC for those who are elements together better.
looking for more details and have enough
rendering power. I hope you have enjoyed these tutorials and that
you will find them useful for your future realistic
If you encounter a long rendering hang, make portrait creations. Thanks for reading!
sure to increase the Dynamic Memory limit to
more than its default 500 MB.
If you are looking for some more control over
your renderings, make sure to explore V-Ray’s
Render Elements, which will provide you with
the ability to render different passes like Diffuse,
Reflections, SSS and so on.
After you render all the necessary passes you
still have some room for improvement in post-
processing. One of the most obvious additions
in post-processing is a background and it’s
very important to match the background with
the lighting. If you have used an image to light
your character, then you could use the same
image as the back plate to perfectly match the
background with the lighting mood.
page 36 Chapter 06
Anto Juricic
Web: http://anto-toni.cgsociety.org/gallery/
Email: monty.band@gmail.com
Even some of the most modern forms of art are influenced by traditional techniques, methods and styles. With
the arrival of ZBrush came an opportunity to put into practice many of the techniques that were used in classical
sculpture by sculptors from some of the most recognizable periods in world history. In this tutorial series Rafael
Ghencev will dissect the history of Greek and Roman sculpture and show you how to create an image in that
style. Not only will Rafael talk you through the sculpting, but will also show how to texture and present your sculpt
in a classical style. This series will also give some great anatomy tips, and provide you with some great tricks to
help you present your sculpts.

Original Author: 3DTotal.com Ltd | Platform: ZBrush | Format: DOWNLOAD ONLY PDF | Pages: 30

Visit 3dtotal.com/shop to see our full range of training products |

Creating cartoon animals is a great starting point for anyone who is thinking about having a go at 3D. Cartoon
animals are usually made up of shapes and forms that are fairly simple to model, and since realism is not
the focus of this kind of work, it gives you an opportunity to be expressive and experimental at all points of
the creative process. In this eBook the artists will be talking you through how to approach this task, firstly by
concentrating on how to create your idea, but then moving on to the modeling, texturing and post-production.
Each chapter provides you with an opportunity to see into the workflow of these outstanding industry
professionals and learn valuable techniques you can use in your own workflow.

Original Author: 3DTotal.com Ltd | Platform: 3DS Max & ZBrush | Format: DOWNLOAD ONLY PDF | Pages: 52

Visit 3dtotal.com/shop to see our full range of training products |