Computer Graphics  
Lecture 032D and 3D Transformations  
Edirlei Soares de Lima  
<edirlei.lima@universidadeeuropeia.pt>  
2
D Transformations  
Transformations are a fundamental part of computer graphics.  
They can be used to position objects, shape objects, change  
viewing positions, and even to change how something is  
viewed (projection transformation).  
Let’s start with 2D transformations: translation, scaling and  
rotation.  
Representation of Points/Objects  
2D objects are represented by a set of points (vertices), {p1,  
p2, ..., pn}, and an associated set of edges {e1, e2, ..., em},  
where each edge is a pair of points e = {pi, pj}.  
We can use Unity to create a mesh with a triangle (ignoring  
the z coordinate).  
...  
private Mesh mesh;  
private Vector3[] vertices;  
void Start(){  
mesh = new Mesh();  
GetComponent<MeshFilter>().mesh = mesh;  
mesh.name = "MyMesh";  
vertices = new Vector3[3];  
vertices[0] = new Vector3(3, 0, 1);  
vertices[1] = new Vector3(5, 0, 1);  
vertices[2] = new Vector3(4, 2, 1);  
mesh.vertices = vertices;  
int[] triangles = new int[3];  
triangles[0] = 0;  
triangles[1] = 2;  
triangles[2] = 1;  
mesh.triangles = triangles;  
}
...  
2
D Translation  
A translation moves an object to a different position on the  
screen. A point can be translated by adding a translation  
coordinate (tx, ty) to the original coordinate (X, Y) to get the  
new coordinate (X’, Y’).  
′  
= ꢃ +ꢄ  
= ꢀ +ꢂ  
2
D Translation  
void Translate2D(float tx, float ty)  
{
for (int i = 0; i < vertices.Length; i++)  
{
vertices[i] = new Vector3(vertices[i].x + tx,  
vertices[i].y + ty);  
}
mesh.vertices = vertices;  
}
void Update()  
{
Translate2D(2 * Time.deltaTime, 0);  
}
2
D Scaling  
Scaling can be achieved by multiplying the original  
coordinates of the object with the scaling factor (Sx, Sy) . The  
object can be either expanded compressed.  
′  
= ×ꢄ  
= ꢀ ×ꢂ  
2
D Scaling  
void Scale2D(float tx, float ty)  
{
for (int i = 0; i < vertices.Length; i++)  
{
vertices[i] = new Vector3(vertices[i].x * sx,  
vertices[i].y * sy);  
}
mesh.vertices = vertices;  
}
void Update()  
{
Scale2D(1.01f, 1);  
}
2
D Scaling Matrix  
We can also represent the scaling transformation using matrix  
multiplication:  
′  
= ×ꢄ  
= ×ꢂ  
0  
0 ꢅꢄ  
ꢃ´  
´ =  
×  
How it works?  
ꢂ  
0
0
ꢄ  
ꢀ × + × 0  
=
× 0 + ×ꢄ  
×  
2
D Scaling Matrix  
void Scale2DMat(float sx, float sy)  
{
float[,] mat = new float[2, 2];  
mat[0, 0] = sx; mat[0, 1] = 0;  
mat[1, 0] = 0; mat[1, 1] = sy;  
for (int i = 0; i < vertices.Length; i++)  
{
vertices[i] = multiply(mat, vertices[i]);  
}
mesh.vertices = vertices;  
}
void Update()  
{
Scale2DMat(1.01f, 1);  
}
Point × Matrix  
Vector3 multiply(float[,] matrix, Vector3 point)  
{
Vector3 result = new Vector3();  
for (int r = 0; r < matrix.GetLength(0); r++)  
{
float s = 0;  
for (int z = 0; z < matrix.GetLength(1); z++)  
s += matrix[r, z] * point[z];  
result[r] = s;  
}
return result;  
}
2
D Rotation  
In rotation, we rotate the object at particular angle θ (theta)  
from its origin.  
′  
= ꢀ × sin + ꢃ × ꢊꢋꢇ(ꢆ)  
= ꢀ × cos ꢃ × ꢇꢈꢉ(ꢆ)  
2
D Rotation  
void Rotate2D(float tx, float ty)  
{
for (int i = 0; i < vertices.Length; i++)  
{
vertices[i] = new Vector3(vertices[i].x * Mathf.Cos(angle)  
-
vertices[i].y * Mathf.Sin(angle),  
vertices[i].x * Mathf.Sin(angle)  
vertices[i].y * Mathf.Cos(angle));  
+
}
mesh.vertices = vertices;  
}
void Update()  
{
Rotate2D(20 * Mathf.Deg2Rad * Time.deltaTime);  
}
2
D Rotation Matrix  
We can also represent rotation transformations using matrix  
multiplication:  
′  
= ꢀ × cos ꢃ × ꢇꢈꢉ(ꢆ)  
ꢃ´  
´ =  
ꢊꢋꢇ ꢆ ꢇꢈꢉ ꢆ  
× ꢇꢈꢉ ꢊꢋꢇ ꢆ  
= ꢀ × sin + ꢃ × ꢊꢋꢇ(ꢆ)  
2
D Rotation Matrix  
void Rotate2DMat(float angle)  
{
float[,] mat = new float[2, 2];  
mat[0, 0] = Mathf.Cos(angle); mat[0, 1] = -Mathf.Sin(angle);  
mat[1, 0] = Mathf.Sin(angle); mat[1, 1] = Mathf.Cos(angle);  
for (int i = 0; i < vertices.Length; i++)  
{
vertices[i] = multiply(mat, vertices[i]);  
}
mesh.vertices = vertices;  
}
void Update()  
{
Rotate2DMat(20 * Mathf.Deg2Rad * Time.deltaTime);  
}
2
D Translation Matrix  
Can we represent translation using matrix multiplication?  
′  
= ꢀ +ꢂ  
= ꢃ +ꢄ  
ꢌ ꢎ  
We need a 2 by 2 matrix, , such that ꢀ × ꢌ + ꢍ × ꢊ = +.  
There is no way to obtain the term!  
It’s impossible to represent translation using a 2 by 2 matrix!  
2
D Homogeneous Coordinates  
What is required at this point is to change the setting (2D  
coordinate space) in which we phrased our original problem.  
In geometry, a common strategy to solve a problem in n space consists  
in rephrasing it to a n+1 space.  
ꢀ´  
Transformation´ ꢀ´  
ꢃ´  
1
1
In other words, the homogeneous coordinate system adds an  
extra virtual dimension. Thus 2D homogeneous coordinates  
are actually 3D. This kind of transformation is called an affine  
transformation.  
2
D Homogeneous Coordinates  
When using homogeneous coordinates to represent  
transformations it is important to distinguish vectors that  
represent positions and directions.  
1
0
1
1
0 is a position.  
1 is a direction.  
1
0
This is important because vectors that represent directions or  
offsets should not change when we translate an object.  
If there is a scaling/rotation transformation in the upper-left 2 × 2  
entries of the matrix, it will apply to the vector, but the translation still  
multiplies with the zero and is ignored.  
2
D Translation Homogeneous Coordinates  
Now we can represent translations as a matrix multiplication:  
1
0 ꢂ  
ꢀ´  
ꢃ´ = × 0 1ꢄ  
1
1
0 0 1  
void Translate2DMatHM(float tx, float ty){  
float[,] mat = new float[3, 3];  
mat[0, 0] = 1; mat[0, 1] = 0; mat[0, 2] = tx;  
mat[1, 0] = 0; mat[1, 1] = 1; mat[1, 2] = ty;  
mat[2, 0] = 0; mat[2, 1] = 0; mat[2, 2] = 1;  
for (int i = 0; i < vertices.Length; i++){  
vertices[i] = multiply(mat, vertices[i]);  
}
mesh.vertices = vertices;  
}
2
D Scale Homogeneous Coordinates  
We can also represent scale as a matrix multiplication in  
homogeneous coordinates:  
0 0  
ꢀ´  
ꢃ´ = × 0 0  
1
1
0 0 1  
void Scale2DMatHM(float sx, float sy){  
float[,] mat = new float[3, 3];  
mat[0, 0] = sx; mat[0, 1] = 0; mat[0, 2] = 0;  
mat[1, 0] = 0; mat[1, 1] = sy; mat[1, 2] = 0;  
mat[2, 0] = 0; mat[2, 1] = 0; mat[2, 2] = 1;  
for (int i = 0; i < vertices.Length; i++){  
vertices[i] = multiply(mat, vertices[i]);  
}
mesh.vertices = vertices;  
}
2
D Rotation Homogeneous Coordinates  
We can also represent rotation as a matrix multiplication in  
homogeneous coordinates:  
ꢀ´  
ꢊꢋꢇ ꢆ ꢇꢈꢉ ꢆ 0  
ꢃ´ = × ꢇꢈꢉ ꢊꢋꢇ 0  
1
1
0
0
1
void Rotate2DMatHM(float angle){  
float[,] mat = new float[3, 3];  
mat[0,0]=Mathf.Cos(angle); mat[0,1]=-Mathf.Sin(angle); mat[0,2]=0;  
mat[1,0]=Mathf.Sin(angle); mat[1,1]=Mathf.Cos(angle); mat[1,2]=0;  
mat[2,0]=0;  
mat[2,1]=0;  
mat[2,2]=1;  
for (int i = 0; i < vertices.Length; i++){  
vertices[i] = multiply(mat, vertices[i]);  
}
mesh.vertices = vertices;  
}
Combining Transformations  
Using homogeneous coordinates, every rotation, translation,  
and scaling operation can be represented by a matrix  
multiplication.  
The advantages of the homogeneous coordinates became  
clear when we need to combine more than one  
transformation at once.  
Applying transformation matrices in sequence is the same as applying  
the product of those matrices once.  
This is a key concept that underlies most graphics hardware  
and software.  
Combining Transformations Example 1  
For example, suppose we wish to first rotate an object 90  
degrees and then scale the object by 2 along the x axis.  
ꢊꢋꢇ ꢆ ꢇꢈꢉ ꢆ 0  
0 0  
rotate(90) = ꢇꢈꢉ ꢊꢋꢇ 0  
scale(2,1) = 0 0  
0
0
1
0
0 1  
0
−1 0  
rotate(90) = 1 0 0  
0 1  
2 0 0  
scale(2,1) = 0 1 0  
0 0 1  
0
When combining transformations, it is very important to remember that  
matrix multiplication is not commutative. So the order of transforms  
does matter. The transforms are applied from the right side first.  
2
0
0
0 0 0 −2 0  
1 0 × 1 0 0 = 1 0 0  
0 1 0 0 1 0 0 1  
0 −1 0  
Combining Transformations Example 1  
0
−2 0  
ꢐꢋꢁꢌꢁꢑ 90 + ꢇꢊꢌꢒꢑ(2,1) = 1 0 0  
0 1  
Square object:  
P3  
P4  
0
0
0 −2 0 0  
ꢓ1 = 0 × 1 0 0 = 0  
P1  
P2  
1
0 0 1  
0 −2 0  
1
1
0
0
ꢓ2 = 0 × 1 0 0 = 10  
1
0
0 0 1  
0 −2 0  
1
−20  
Rotated and scaled square:  
ꢓ3 = 10 × 1 0 0 = 0  
P4  
P2  
1
0 0 1  
0 −2 0  
1
1
0
−20  
ꢓ4 = 10 × 1 0 0 = 10  
1
0 0 1  
1
P3  
P1  
Combining Transformations Example 1  
float[,] multiply(float[,] matrix1, float[,] matrix2)  
{
float[,] result = new float[matrix1.GetLength(0),  
matrix2.GetLength(1)];  
for (int i = 0; i < matrix1.GetLength(0); i++)  
{
for (int j = 0; j < matrix2.GetLength(1); j++)  
{
float s = 0;  
for (int k = 0; k < matrix1.GetLength(1); k++)  
{
s += matrix1[i, k] * matrix2[k, j];  
}
result[i, j] = s;  
}
}
return result;  
}
Combining Transformations Example 1  
void RotateScale(float angle, float sx, float sy)  
{
float[,] rmat = new float[3, 3];  
rmat[0,0]=Mathf.Cos(angle);rmat[0,1]=-Mathf.Sin(angle);rmat[0,2]=0;  
rmat[1,0]=Mathf.Sin(angle);rmat[1,1]=Mathf.Cos(angle); rmat[1,2]=0;  
rmat[2,0]=0;  
rmat[2,1]=0;  
rmat[2,2]=1;  
float[,] smat = new float[3, 3];  
smat[0,0] = sx; smat[0,1] = 0; smat[0,2] = 0;  
smat[1,0] = 0; smat[1,1] = sy; smat[1,2] = 0;  
smat[2,0] = 0; smat[2,1] = 0; smat[2,2] = 1;  
float[,] finalmat = multiply(scalemat, rmat);  
for (int i = 0; i < vertices.Length; i++)  
{
vertices[i] = multiply(finalmat, vertices[i]);  
}
mesh.vertices = vertices;  
}
Combining Transformations Example 2  
In order to rotate an object around a specific point (such as its  
center or a corner). We need to perform 3 operations:  
(1) translate the object to the origin;  
(2) rotate the object;  
(3) translate the object back to its initial position.  
Suppose we wish to rotate our object 90 degrees around its  
upper right corner (P4).  
P3  
P4  
P1  
P2  
Combining Transformations Example 2  
Rotation of 90 degrees around the upper right corner (P4 = (10, 10)):  
1
0 ꢁꢂ  
= 0 1 ꢁ = 0 1 −10  
0 1  
1
0 −10  
P3  
P1  
P4  
P2  
0
0
0 1  
ꢊꢋꢇ ꢆ ꢇꢈꢉ ꢆ 0  
ꢖ = ꢇꢈꢉ ꢊꢋꢇ 0 = 1 0 0  
0 −1 0  
0
0
1
0 0 1  
1
0 ꢁꢂ  
= 0 1 ꢁ = 0 1 10  
0 1  
1
0 10  
0
0
0 1  
1
0 10 0 −1 20  
× × = 0 1 10 × 1 0 0 × 0 1 −10 = 1 0 0  
0 1 0 0 1 0 0 1 0 0 1  
0 −1 0  
1 0 −10  
0
Combining Transformations Example 2  
Rotation of 90 degrees around the upper right corner (P4 = (10, 10)):  
Square object:  
0
−1 20  
× × = 1 0 0  
0 1  
P3  
P1  
P4  
P2  
0
0
0 −1 20 20  
ꢓ1 = 0 × 1 0 0 = 0  
0 0 1  
0 −1 20  
1
1
1
0
20  
ꢓ2 = 0 × 1 0 0 = 10  
1
0
ꢓ3 = 10 × 1 0 0 = 0  
0 0 1  
0 −1 20  
1
10  
Rotated square:  
P4 P2  
1
0 0 1  
0 −1 20  
1
10  
1
0
P3 P1  
ꢓ4 = 10 × 1 0 0 = 10  
1
0 0 1  
1
Combining Transformations Example 2  
void RotateAroundPoint(float angle, float px, float py)  
{
float[,] rmat = new float[3, 3];  
rmat[0,0]=Mathf.Cos(angle);rmat[0,1]=-Mathf.Sin(angle);rmat[0,2]=0;  
rmat[1,0]=Mathf.Sin(angle);rmat[1,1]=Mathf.Cos(angle); rmat[1,2]=0;  
rmat[2,0]=0;  
rmat[2,1]=0;  
rmat[2,2]=1;  
float[,] tran1mat = new float[3, 3];  
t1mat[0, 0] = 1; t1mat[0, 1] = 0; t1mat[0, 2] = -px;  
t1mat[1, 0] = 0; t1mat[1, 1] = 1; t1mat[1, 2] = -py;  
t1mat[2, 0] = 0; t1mat[2, 1] = 0; t1mat[2, 2] = 1;  
float[,] tran2mat = new float[3, 3];  
t2mat[0, 0] = 1; t2mat[0, 1] = 0; t2mat[0, 2] = px;  
t2mat[1, 0] = 0; t2mat[1, 1] = 1; t2mat[1, 2] = py;  
t2mat[2, 0] = 0; t2mat[2, 1] = 0; t2mat[2, 2] = 1;  
float[,] finalmat = multiply(multiply(t2mat, rmat), t1mat);  
for (int i = 0; i < vertices.Length; i++)  
{
vertices[i] = multiply(finalmat, vertices[i]);  
}
mesh.vertices = vertices;  
}
Exercise 1  
1
) Reflection is a transformation that flips the object over the  
axis of reflection (x or y).  
Reflection over the x-axis:  
a) What is the reflection matrix  
for the x-axis?  
b) What is the reflection matrix  
for the y-axis?  
Reflection over the y-axis:  
c) Implement and apply the  
reflection transformation.  
Exercise 2  
2
) Shearing is a transformation that displaces each point in a  
fixed direction, by an amount proportional to its signed  
distance from a line that is parallel to that direction.  
Shear over the x-axis:  
a) What is the shearing matrix  
for the x-axis?  
b) What is the shearing matrix  
for the y-axis?  
Shear over the y-axis:  
c) Implement and apply the  
shearing transformation.  
3
D Transformations  
3D Transformations are an extension of the 2D transformations.  
We can also use homogeneous coordinates in 3D by adding a  
fourth coordinate to the transformation matrices.  
Representation of 3D Objects  
3D objects are represented by a set of triangles. Each triangle  
is defined by three vertices. In addition, each vertex has a  
normal vector, which usually points outward (perpendicular  
to the mesh surface).  
The vertices and triangles define the spatial structure of the object.  
The normals define how light affects the surface of the object.  
vertices = new Vector3[18];  
/back triangle vertices  
/
vertices[0] = new Vector3(-10, 0, -10);  
vertices[1] = new Vector3(10, 0, -10);  
vertices[2] = new Vector3(0, 10, 0);  
//front triangle vertices  
vertices[3] = new Vector3(-10, 0, 10);  
vertices[4] = new Vector3(10, 0, 10);  
vertices[5] = new Vector3(0, 10, 0);  
//left triangle vertices  
vertices[6] = new Vector3(-10, 0, 10);  
vertices[7] = new Vector3(-10, 0, -10);  
vertices[8] = new Vector3(0, 10, 0);  
//right triangle vertices  
vertices[9] = new Vector3(10, 0, 10);  
vertices[10] = new Vector3(10, 0, -10);  
vertices[11] = new Vector3(0, 10, 0);  
//botton triangle 1 vertices  
vertices[12] = new Vector3(-10, 0, -10);  
vertices[13] = new Vector3(10, 0, -10);  
vertices[14] = new Vector3(-10, 0, 10);  
//botton triangle 2 vertices  
vertices[15] = new Vector3(10, 0, 10);  
vertices[16] = new Vector3(10, 0, -10);  
vertices[17] = new Vector3(-10, 0, 10);  
int[] triangles = new int[18];  
/back triangle  
/
triangles[0] = 0;  
triangles[1] = 2;  
triangles[2] = 1;  
//front triangle  
triangles[3] = 4;  
triangles[4] = 5;  
triangles[5] = 3;  
//left triangle  
triangles[6] = 6;  
triangles[7] = 8;  
triangles[8] = 7;  
//right triangle  
triangles[9] = 10;  
triangles[10] = 11;  
triangles[11] = 9;  
//botton triangle 1  
triangles[12] = 13;  
triangles[13] = 14;  
triangles[14] = 12;  
//botton triangle 2  
triangles[15] = 15;  
triangles[16] = 17;  
triangles[17] = 16;  
Vector3[] normals = new Vector3[18];  
normals[0] = Vector3.back + Vector3.up;  
normals[1] = Vector3.back + Vector3.up;  
normals[2] = Vector3.back + Vector3.up;  
normals[3] = Vector3.forward + Vector3.up;  
normals[4] = Vector3.forward + Vector3.up;  
normals[5] = Vector3.forward + Vector3.up;  
normals[6] = Vector3.left + Vector3.up;  
normals[7] = Vector3.left + Vector3.up;  
normals[8] = Vector3.left + Vector3.up;  
normals[9] = Vector3.right + Vector3.up;  
normals[10] = Vector3.right + Vector3.up;  
normals[11] = Vector3.right + Vector3.up;  
normals[12] = Vector3.down;  
normals[13] = Vector3.down;  
normals[14] = Vector3.down;  
normals[15] = Vector3.down;  
normals[16] = Vector3.down;  
normals[17] = Vector3.down;  
3
D Translation Homogeneous Coordinates  
Using homogeneous coordinates, we can also represent 3D  
translation as a matrix multiplication:  
1
0 0 ꢂ  
ꢀ´  
0
ꢄ  
ꢃ´  
ꢘ´  
1
1
0 1  
0
0
=
×
0 1 ꢁꢙ  
0 0 1  
void Translate3D(float tx, float ty, float tz){  
Matrix4x4 translation_matrix = new Matrix4x4();  
translation_matrix.SetRow(0, new Vector4(1f, 0f, 0f, tx));  
translation_matrix.SetRow(1, new Vector4(0f, 1f, 0f, ty));  
translation_matrix.SetRow(2, new Vector4(0f, 0f, 1f, tz));  
translation_matrix.SetRow(3, new Vector4(0f, 0f, 0f, 1f));  
for (int i = 0; i < vertices.Length; i++){  
vertices[i]=translation_matrix.MultiplyPoint(vertices[i]);  
}
mesh.vertices = vertices;  
}
3
D Scale Homogeneous Coordinates  
We can also represent 3D scale as a matrix multiplication:  
0 0 0  
ꢀ´  
0
0 0  
0 ꢅ 0  
0 0 1  
ꢃ´  
ꢘ´  
1
1
=
×
0
0
void Scale3D(float sx, float sy, float sz){  
Matrix4x4 scale_matrix = new Matrix4x4();  
scale_matrix.SetRow(0, new Vector4(sx, 0f, 0f, 0));  
scale_matrix.SetRow(1, new Vector4(0f, sy, 0f, 0));  
scale_matrix.SetRow(2, new Vector4(0f, 0f, sz, 0));  
scale_matrix.SetRow(3, new Vector4(0f, 0f, 0f, 1f));  
for (int i = 0; i < vertices.Length; i++){  
vertices[i]=scale_matrix.MultiplyPoint(vertices[i]);  
}
mesh.vertices = vertices;  
}
3
D Rotation Homogeneous Coordinates  
3D rotations are not exactly same as 2D rotations. In 3D, we  
have to specify the angle of rotation along with the axis of  
rotation (we can perform 3D rotations about X, Y, and Z axes).  
1
0
0
0
0
ꢊꢋꢇ θ ꢇꢈꢉ ꢆ 0  
ꢇꢈꢉ ꢆ ꢊꢋꢇ 0  
0
0
=  
=  
=  
0
0
1
ꢊꢋꢇ θ 0 ꢇꢈꢉ 0  
0
1
0
0
ꢇꢈꢉ ꢆ 0 ꢊꢋꢇ 0  
0
0
0
1
0
0
1
ꢊꢋꢇ θ −ꢇꢈꢉ ꢆ  
ꢇꢈꢉ ꢆ ꢊꢋꢇ ꢆ  
0
0
0
0
0
0
0
0
1
3
D Rotation Homogeneous Coordinates  
1
0
0
0
0
ꢊꢋꢇ θ ꢇꢈꢉ ꢆ 0  
ꢇꢈꢉ ꢆ ꢊꢋꢇ 0  
0
0
=  
0
0
1
void RotateX3D(float angle)  
{
Matrix4x4 rxmat = new Matrix4x4();  
rxmat.SetRow(0,new Vector4(1f,0f,  
0f,  
0f));  
rxmat.SetRow(1,new Vector4(0f,Mathf.Cos(angle),-Mathf.Sin(angle),0f));  
rxmat.SetRow(2,new Vector4(0f,Mathf.Sin(angle),Mathf.Cos(angle), 0f));  
rxmat.SetRow(3,new Vector4(0f,0f,  
0f,  
1f));  
for (int i = 0; i < vertices.Length; i++){  
vertices[i] = rxmat.MultiplyPoint(vertices[i]);  
normals[i] = rxmat.MultiplyPoint(normals[i]);  
}
mesh.vertices = vertices;  
mesh.normals = normals;  
}
3
D Rotation Homogeneous Coordinates  
ꢊꢋꢇ θ 0 ꢇꢈꢉ 0  
0
ꢇꢈꢉ ꢆ 0 ꢊꢋꢇ 0  
1
0
0
=  
0
0
0
1
void RotateY3D(float angle)  
{
Matrix4x4 rymat = new Matrix4x4();  
rymat.SetRow(0,new Vector4(Mathf.Cos(angle), 0f,Mathf.Sin(angle),0f));  
rymat.SetRow(1,new Vector4(0f, 1f,0f, 0f));  
rymat.SetRow(2,new Vector4(-Mathf.Sin(angle),0f,Mathf.Cos(angle),0f));  
rymat.SetRow(3,new Vector4(0f,  
0f,0f,  
1f));  
for (int i = 0; i < vertices.Length; i++){  
vertices[i] = rymat.MultiplyPoint(vertices[i]);  
normals[i] = rymat.MultiplyPoint(normals[i]);  
}
mesh.vertices = vertices;  
mesh.normals = normals;  
}
3
D Rotation Homogeneous Coordinates  
ꢊꢋꢇ θ −ꢇꢈꢉ ꢆ  
ꢇꢈꢉ ꢆ ꢊꢋꢇ ꢆ  
0
0
1
0
0
0
=  
0
0
0
0
0
1
void RotateZ3D(float angle)  
{
Matrix4x4 rzmat = new Matrix4x4();  
rzmat.SetRow(0,new Vector4(Mathf.Cos(angle),-Mathf.Sin(angle),0f,0f));  
rzmat.SetRow(1,new Vector4(Mathf.Sin(angle), Mathf.Cos(angle),0f,0f));  
rzmat.SetRow(2,new Vector4(0f,  
rzmat.SetRow(3,new Vector4(0f,  
0f,  
0f,  
1f,0f));  
0f,1f));  
for (int i = 0; i < vertices.Length; i++){  
vertices[i] = rzmat.MultiplyPoint(vertices[i]);  
normals[i] = rzmat.MultiplyPoint(normals[i]);  
}
mesh.vertices = vertices;  
mesh.normals = normals;  
}
3
D Rotation Homogeneous Coordinates  
When we rotate a 3D object, we also need to apply the same  
rotation to the normal vectors of the object:  
...  
for (int i = 0; i < vertices.Length; i++){  
vertices[i] = rxmat.MultiplyPoint(vertices[i]);  
normals[i] = rxmat.MultiplyPoint(normals[i]);  
}
mesh.vertices = vertices;  
mesh.normals = normals;  
...  
We can visualize the normals in the Unity editor:  
private void OnDrawGizmos(){  
if (vertices == null) return;  
for (int i = 0; i < vertices.Length; i++){  
Gizmos.color = Color.yellow;  
Gizmos.DrawRay(vertices[i], normals[i]);  
}
}
Exercise 3  
3
) Rotations in 3D are performed around the origin of the  
coordinates system (0, 0, 0). How can we rotate a 3D  
object around a specific position?  
Implement in Unity the functions to rotate objects around  
specific positions (such as the center of the object) in the X,  
Y, and Z axes.  
Homogeneous Coordinates  
Homogeneous coordinates are used nearly universally to  
represent transformations in graphics systems.  
They underlie the design and operation of renderers implemented in  
all graphics hardware.  
Homogeneous coordinates not only clean up the code for  
transformations, but they also simplify the composition of  
complex transformations.  
Further Reading  
Hughes, J. F., et al. (2013). Computer Graphics: Principles and Practice  
(
0
3rd ed.). Upper Saddle River, NJ: Addison-Wesley Professional. ISBN: 978-  
-321-39952-6.  
Chapter 10: Transformations in Two Dimensions  
Chapter 11: Transformations in Three Dimensions  
Marschner, S., et al. (2015). Fundamentals of Computer Graphics (4th  
ed.). A K Peters/CRC Press. ISBN: 978-1482229394.  
Chapter 6: Transformation Matrices