3D Transformation

o Instead of creating every o Transformations are

new object from scratch, measured in a coordinate
we may transform some matrix
existing objects o There are two possible ways
of specifying the Z-axis, which
o We also need to transform gives rise to a left-handed or
objects from one space to a right-handed system
another, e.g., world space
to view space/camera
space
o Instead of performing each
transformation alone, we
accumulate the
transformations in a matrix
before applying them to
objects
 2D Transformation
o Scaling so
o X’ = aX x' = x cos f - y sin f
o Y’ = aY y' = y cos f + x sin f
o Translation
o X’=x+b  x'   cos  sin   x 
o Y’=y+b
     
 y'   cos sin   y 
o Rotation
x' = r cos (q + f ) = y
r cos q cos f - r sin q sin f

y' = r sin (q + f ) =
r sin q cosf + r cos q sin f x = r cos q
x y = r sin q
 Homogeneous Coordinates
o In 2D, three elements can be used to represent a point, e.g.,
(x,y,1) to represent (x,y) or (wx,wy,w) w0 to represent (x,y)
o (x,y,w) is a homogeneous coordinate representing the point
(x/w, y/w)
o A point has many homogeneous representations in
homogeneous coordinates.
o Reason for homogeneous coordinates
o Allow an affine transformation to be represented by a single
matrix
o Useful for perspective transformation
o Allow bezier curves and sline curves to represent circles and
other conic sections
 2D Transformation Matrices

o Translation matrix: 1 0 bx 
0 1 b 
 y

0 0 1 
o Scaling matrix:  s x 0 0
 0 s 0
 y 
 0 0 1

o Rotation matrix: cos q  sin q 0

 sin q cos q 0

 0 0 1
 3D Translation and Scaling

o In 3D four elements can be used to represent a

point so the translation and scaling
transformations may be represented as:

1 0 0 trX
Translation 0 1 0 trY
matrix 0 0 1 trZ
0 0 0 1

Scaling
SX 0 0 0
matrix 0 SY 0 0
0 0 SZ 0
0 0 0 1
 3D Rotation

o In 2D, rotation is about a point

o In 3D, rotation is about a vector, which can be
done through rotations about x, y or z axes
o Positive rotations are anti-clockwise, negative
rotations are clockwise, when looking down a
positive axis towards the origin
y y
y

x x
x

z z z
 Major Axis Rotation Matrices

1 0 0 0
0 cosq  sinq 0
o about X axis Rx   
0 sinq cosq 0
 
0 0 0 1

 cosq 0 sinq 0
o about Y axis  0 1 0 0
Ry   
 sinq 0 cosq 0
 
 0 0 0 1

o about Z axis cosq  sinq 0 0

 sinq cosq 0 0
Rz   
 0 0 1 0
 
 0 0 0 1
 Rotation Axis

o In general rotation vector does not pass through

coordinate origin

The idea is to turn rotation about an arbitrary axis to

rotations about coordinate axes
 Rotation about an Arbitrary Axis
o We will first rotate about x then about y to align the arbitrary
axis with z axis, then do the required rotation about z, then
undo rotations about y and x axes a
sin  
k
c
cos  
k
k is length of Uxz
c

 Uyz is proj. of U on yz plane xz=(a,0,c)

 d is length of Uyz k is length of Uxz

(1)

Reverse translation and rotations abut x and y to move the arbitrary

vector U back to its initial position (leave only required rotation about U)
 Rotation about an Arbitrary Axis

c a 
 0  0 Why?
k k  Cos it’s
0 0 0 0
negative
a 0
c
0 rotation
k k 
0 1 
about y
 0 0
about x axis about y axis

1 0 0 0  cosq 0 sinq 0 cosq  sinq 0 0

 0  0
0 cosq  sinq 0 1 0 0 R   sinq cosq 0

Rx    Ry   
 0 0
z
0 sinq cosq 0  sinq 0 cosq 0 0 1
     
0 0 0 1  0 0 0 1  0 0 0 1
 Rotation about an Arbitrary Axis
o Alternatively, we can first rotate about y then about x to align
the arbitrary axis with z axis, then do the required rotation
about z, then undo unwanted rotations
- rotate about y axis by a to bring vector n to YoZ plane
Y
- rotate about x axis by  to align n with Z axis

nx ny
a  tan 1
  tan 1

nz d
o n(nx, ny, nz) X
  d  nx + nz
2 2

a d Final transformation matrix can be

obtained by changing x to y and y to
Z x in equation (1)
(1)
 Transformation in OpenGL

o Because of the way OpenGL applies the

transformations to our matrices, the last
transformation command in your program is the
first one applied to the vertices
o The order of transformations is critical. If you do
transformation A and then transformation B, you
almost certainly get something different than if
you do them in the opposite order
 Exercise - Rotate by 45 about x axis

o To rotate by 45 degrees about x-axis, use the

following matrix:

1 0 0 0
1 0 0 0  1 1 
0 
cos 45  sin 45 0 
0 0
R x (45)  
2 2
 
0 sin 45 cos 45 0 0 1 1
0
 
0 0 0 1  2 2 
0 0 0 1
 Exercise - Translate by (1, 1, 1)

o Translation by (tx, ty, tz) is o So translation by (1, 1, 1) is

achieved using the the following:
following matrix:

1 0 0 tx  1 0 0 1
0 0 t y  0
M T (t x , t y , t z )  
1 1 0 1
0 1 tz  M T (1,1,1)  
0 0 0 1 1
   
0 0 0 1
0 0 0 1
 Exercise - Rotate by 45 about axis (1, 1, 1)

o Rotation by 45 degrees about (1, 1, 1) can be

achieved by a few successive rotations about the
major axes, which can be represented as a single
composite transformation
1 n x n
Y a  tan   tan 1 y
nz d
d  nx + nz
2 2

o n(nx, ny, nz) X d  nx + nz  2  1.414

2 2

 
nx = 1
a d   tan 1
ny
 tan 1
1
 35.264
SO d
ny = 1 2
n 1
Z
nz = 1 a  tan 1 x  tan 1  45
nz 1
 Q3 - Arbitrary Axis Rotation
o The composite
transformation can then be
represented as:
M R (1,1,1)  Ry (a )  Rx (  )  Rz (q )  Rx (  )  Ry (a )

 cos( 45) 0 sin( 45) 0 1 0 0 0 cos( 45)  sin( 45) 0 0

 0 0 cos(35.2)  sin( 35.2)
 0 1 0 0  sin( 45) cos( 45) 0 0
=  sin( 45) 0 cos(45) 0

0 sin( 35.2) cos(35.2) 0  0 0 1 0
    
 0 0 0 1 0 0 0 1  0 0 0 1

1 0 0 0  cos( 45) 0 sin( 45) 0  x

0 cos(35.2)  sin( 35.2) 0  0 1 0 0   
    y
0 sin( 35.2) cos(35.2) 
0  sin( 45) 0 cos( 45) 0 z
 
0 0 0 1  0 0 0 1

 
1
neg. rotation about Y
 Exercise
o Derive an expression for the o Question: What’s q , and
matrix transformation that rotates
rotation around which
an object by 45 degrees about a
vector, originating from P1(0,1,0) axis?
on the XoY plane and passing
through point P2(1,0,0). You do
NOT need to calculate matrix y
multiplication, but show the
components of the transformation P1(0,1,0)
matrix
o P1 and P2 are on XoY plane, so
vector P1P2 is on XoY plane P2(1,0,0)
o x
o After translation to the origin q
vector P1’P2’ is still on the XoY
plane, and can be rotated by q z
to align with X axis
 Exercise

1 0 0 0  cosq 0 sinq 0 cosq  sinq 0 0

0 cosq 0  0 0  sinq
 sinq 1 0 cosq 0 0
Rx    Ry    Rz   
0 sinq cosq 0  sinq 0 cosq 0  0 0 1 0
     
0 0 0 1  0 0 0 1  0 0 0 1

The rotation transformation around P1P2 is

1 1
1 0 0 0 cos45  sin 45 0 0 1 0 0 0 cos45  sin 45 0 0 1 0 0 0
0  1  sin 45 cos45 0 0 0 cos45  sin 45 0  sin 45 cos45 0 0 

1 0
       0 1 0  1

0 0 1 0  0 0 1 0 0 sin 45 cos45 0  0 0 1 0  0 0 1 0
         
0 0 0 1  0 0 0 1  0 0 0 1  0 0 0 1  0 0 0 1
 Exercise
1 T 1 0 0 0  1 0 0 0
cos 45  sin 45 0 0 cos 45  sin 45 0 0  cos 45 sin 45 0 0 0
 sin 45 cos 45 0  1 0 0
0  sin 45 cos 45 0  sin 45 cos 45 0 0 1 1 0
0 0   
      0 0 1 0  0 0 1 0
 0 0 1 0  0 0 1 0  0 0 1 0   
      0 0 0 1  0 0 0 1
 0 0 0 1  0 0 0 1  0 0 0 1
1 0 0 0 1 0 0 0
  cos 45 sin 45 0 0 cos 45  sin 45 0 0 1 0 0 0  0 0 0  0 1
  sin 45 cos 45  
1

1 0

 0 0  sin 45 cos 45 0 0  0 1 0 0   0 1 0  0 0
    0 0 1
 TEST :  0 0 1 0  0 0 1 0  0 0 1 0     
      0 0 0 1  0 0 0 1
 1 0 0 0 1 
  0 0 0 1  0 0 0
1 0 0 0
1
1 0 0 0
0 1 0  1 0 1 0 1
   
0 0 1 0 0 0 1 0
   
So, finally, the rotation transformation around P1P2 is 0 0 0 1 0 0 0 1

1 0 0 0  cos 45 sin 45 0 0 1 0 0 0cos 45  sin 45 0 0 1 0 0 0

0 1 0 1

 sin 45 cos 45 0 0 0 cos 45  sin 45 0 sin 45 cos 45 0 0

0 1 0  1
      
0 0 1 0  0 0 1 0 0 sin 45 cos 45 0 0 0 1 0 0 0 1 0
    0    
0 0 0 1  0 0 0 1  0 0 1 0 0 0 1 0 0 0 1
A
B C D is the rotation matrix
 Inverse Matrix
o Inverse matrix: if AB=I (identity matrix), then B is called A’s
inverse matrix, written as

B  A , AA  I
1 1

o The inverse matrix of a rotation matrix is the transpose (by

flipping the matrix around diagonal) of the rotation matrix
o The inverse matrix of a (invertible) square matrix M can be
obtained by aligning side by side M and an identity matrix I,
and perform the same row operations on both M and I until
M is reduced to an identity matrix, and I is changed and
becomes the inverse matrix of M
 Inverse of a Square Matrix

o Find the inverse of a 3 x 3 o Multiply row(1) by - 2 and add

square matrix A given by it to row(2). also multiply
row(1) by - 3 and add it to
[1 1 2] row(3).
A = [2 4 -3]
[3 6 -5] [1 1 2 | 1 0 0] row(1)
[0 2 -7 | -2 1 0] row(2)
o Augment the given matrix by [0 3 -11 | -3 0 1] row(3)
the identity matrix then use
basic row operations so that o Multiply row(2) by 3 and
the matrix on the left row(3) by -2, add them and
becomes identify matrix. put the row obtained in
row(3).
[1 1 2 | 1 0 0] row(1)
[2 4 -3 | 0 1 0] row(2) [1 1 2 | 1 0 0] row(1)
[3 6 -5 | 0 0 1] row(3) [0 2 -7 | -2 1 0] row(2)
[0 0 1 | 0 3 -2] row(3)
 Inverse of a Square Matrix
o Multiply row(1) by -2 and add it to o Mow multiply all terms in row(1)
row(2) and put the row obtained by -1/2 and multiply all terms in
in row(1). row(2) by 1/2.
[-2 0 -11 | -4 1 0] row(1)
[ 0 2 -7 | -2 1 0] row(2) [1 0 0 | 2 -17 11] row(1)
[ 0 0 1 | 0 3 -2] row(3) [0 1 0 | -1 11 -7] row(2)
[0 0 1 | 0 3 -2] row(3)
o Multiply row(3) by 11 and add it to
row(1) and put the row obtained o Now that we have the identity
in row(1). matrix on the left, the 3 by 3
matrix on the right is the inverse
[-2 0 0 | -4 34 -22] row(1) matrix:
[0 2 -7 | -2 1 0] row(2)
[0 0 1 | 0 3 -2] row(3) [ 2 -17 11]
[-1 11 -7]
o Multiply row(3) by 7 and add it to [ 0 3 -2]
row(2) and put the row obtained
in row(2).
o A similar method based on row
operations can in principle be
[-2 0 0 | -4 34 -22] row(1) used to find the inverse of any
[ 0 2 0 | -2 22 -14] row(2) invertible (that has an inverse)
[ 0 0 1 | 0 3 -2 ] row(3) square matrix.
there are other ways to invert a matrix..

A = [a b] We now need to solve the first and second

equations simultaneously to find a' and c' in
[c d] terms of a, b, c and d.

Let B be the inverse of A given by Multiply the first equation by d and the
second equation by b and subtract the left
and right terms of the equations obtained to
B = [a' b'] find
[c' d']
a'(a d - b c) = d
From definition, we have AB = I a ' = d / (a d - b c)
[a b] [a' b'] = [1 0] Substitute a' in the second equation to obtain
[c d] [c' d'] [0 1] c'= - c / (a d - b c)

We need to find terms a', b', c' and d' To find b' and d', multiply equation (3) by d
and equation (4) by b and subtract the left
of B (inverse) in terms of a, b, c and d and right terms of the equations obtained to
which are the terms of matrix A. find
Multiply matrices A and B above.
b'(a d - b c) = - b or b' = - b / (a d - b c)
[a a' + b c' a b' + b d'] = [1 0]
[c a' + d c' c b' + d d'] [0 1] Substitute b' in equation (3) to obtain d'= a /
(a d - b c)
The above gives 4 equations Finally matrix B, the inverse of A, is given by

a a' + b c' = 1 (equation 1) B = 1 / (a d - b c) [d -b]

[-c a]
c a' + d c' = 0 (equation 2)
a b' + b d' = 0 (equation 3)
c b' + d d' = 1 (equation 4)
 Inverse of a 2 by 2 Square Matrix

o In solving the question in o Use the above formula for

the above example, we the inverse of a 2 x 2
have found a formula for square matrix to find the
the inverse of any inverse of
invertible 2 x 2 matrix.
o A 2 x 2 matrix will have an A = [3 5]
inverse if its determinant [2 3]
D=ad-bc
is not equal to zero since o According to formula
division by zero is not above the inverse of A is
allowed given by

1 / [ 3*3 – 5*2 ] [ 3 -5 ]
[-2 3 ]
o Simplify to obtain inverse
of A:
[ -3 5 ]
[ 2 -3 ]
 Transformation Exercise
o An object is defined by four vertices A (0,0,0), B (1,0,0), C
(0,1,0) and D (0,0,1). The object has to be rotated by +45
degrees about a line passing through point C (0,1,0) with a
direction V = j + k. Calculate the vertex positions after the
rotation.
z
 1 1 
  0 0 1 0 0 0
2 2  1 1  1 0 0 0
  0  0 
1 1 0 1 0  1
Q 0 0  2 2

 2 2  0 1 1
0  0 0 1 0
D 

0 0 1 0 

2 2  0
1 
0 0 1

 0 0 0 1 0 0 0

V
then left multiply Q with
inverse transformations that
A y move V back to its original position
C to arrive at a matrix M.
then multiply M with each of
B A, B, C, D to get vertex positions
x
after the rotation around V.
 OpenGL Matrix Stacks
o OpenGL has three matrix o Each time you call any geometric
stacks: primitive function, the primitive is
o Model/View transformed by the current matrix
glMatrixMode(GL_MODELVIEW o Say you call glLoadIdentity(), and
); glTranslatef(0,0,10), the current
o Projection matrix stack is
glMatrixMode(GL_PROJECTIO 1 0 0 0 
N);  
o Textures  0 1 0 0 
glMatrixMode(GL_TEXTURE);  0 0 1 10 
 
o OpenGL keeps a
transformation stack, and 0 0 0 1 
matrix at the top of the stack o If you now call glVertex3f(0,0,0) now,
is the current matrix C the vertex will appear at (0,0,10).
o Each matrix call, e.g., o If you need the current matrix for
glRotate, or glTranslate, or later use, you need to save a copy of
glScale creates a it on the stack using glPushMatrix().
transformation matrix M and
right multiplies it with the o When you need to use the saved
current matrix C, resulting in a matrix, you can use glPopMatrix().
new current matrix CM
 I*T

OpenGL Matrix Stacks

I*T
I*T

o The modelling stack in action:

o glLoadIdentity(); I*T*T
o glTranslatef(0,10,0); I*T

o glPushMatrix();
o glTranslatef(10,0,0); I*T*T*R
I*T
o glRotatef(45,0,1,0);
o glPushMatrix();
I*T*T*R
o glRotatef(45,0,1,0);
I*T*T*R
o glPopMatrix(); I*T
o glPopMatrix();
o glPopMatrix(); I*T*T*R*R
I*T*T*R
I*T
Transformation in OpenGL
Pyramid Example
 Pyramid Example
void showcube (double x, double y, o The first cube drawn uses the
double z, double angle) { call:
//Shows a cube at x,y,z rotated by
angle around y axis showcube(-9,0,0,45), which
draws a cube translated by 9 in
glPushMatrix(); the –x direction and rotated 45
glTranslatef(x,y,z); degrees around the y axis.
glRotatef(angle,0,1,0); o This results in the following
colorcube(); transformation matrix:
glPopMatrix();
}
o Before any showcube operation a
copy of the current matrix is
saved onto the matrix stack to
become current matrix.
o After the cube is drawn the
current matrix is popped and the
saved matrix becomes the
current matrix.
 Pyramid Example
o void showpyramid(void)
{ o //Rear o //Rear
o glPushMatrix(); o showcube(-6,0,-3,45); o showcube(-3,3,-3,0);
o o showcube(0,0,-3,45); o showcube(3,3,-3,0);
o //Move to centre of o showcube(6,0,-3,45); o showcube(0,3,-6,0);
bottom row
o showcube(-3,0,-6,45);
o glTranslatef(0,-9,0);
o showcube(3,0,-6,45); o //3rd row
o //Bottom row o showcube(0,0,-9,45);
o //2nd row o showcube(-3,6,0,45);
o //Centre
o showcube(3,6,0,45);
o showcube(-9,0,0,45);
o showcube(-3,0,0,45); o //Centre o showcube(0,6,3,45);
o showcube(3,0,0,45); o showcube(0,6,-3,45);
o showcube(0,3,0,0);
o showcube(9,0,0,45);
o showcube(6,3,0,0); o //Top row
o //Front o showcube(-6,3,0,0); o showcube(0,9,0,0);
o showcube(-6,0,3,45);
o showcube(0,0,3,45);
o //Front o glPopMatrix();
o showcube(6,0,3,45);
o }
o showcube(-3,0,6,45); o showcube(-3,3,3,0);
o showcube(3,0,6,45);
o showcube(3,3,3,0);
o showcube(0,0,9,45);
o showcube(0,3,6,0);
 Christmas Tree Example

o The scene is made of a tree (3

Cones), a pot (a Torus), a star (a
Triangle), and some presents (3
Cubes)
o OpenGL glutSolidCone creates a
cone at the coordinate origin
o We need to rotate the cone so
that it stands up in the positive Z
direction
o We need to translate the cones to
make the tree
o We need translations to draw the
pot using OpenGL glutSolidTorus
and the star using
GL_TRIANGLES
o We need translations to draw the
presents using OpenGL
glutSolidCube
 Christmas Tree Example (3 cones+1 pot)

o void drawScene() { o //Draw bottom cone

o glClear(GL_COLOR_BUFFER_BI o glutSolidCone(1.0, 3.0, 10, 10);
T | GL_DEPTH_BUFFER_BIT); glPushMatrix();
o glMatrixMode(GL_MODELVIEW); o glTranslatef(0.0f, 0.0f, 1.0f);
o glLoadIdentity(); o //Draw middle cone
//Move the scene down 2 and o glutSolidCone(0.75, 2.0, 10, 10);
forward 8 units glPushMatrix();
o glTranslatef(0.0f, -2.0f, -8.0f); o glTranslatef(0.0f, 0.0f, 1.0f);
o //Continually rotate scene about o //Draw top cone
the y-axis o glutSolidCone(0.5, 1.0, 10, 10);
o glRotatef(angle, 0.0f, 1.0f, 0.0f); o glPopMatrix();
o //Save the transformations o glPopMatrix();
performed thus far
o glPushMatrix(); o //Draw a pot
o //Rotate about the x-axis so the o glPushMatrix();
cones are upright
o glRotatef(-90, 1.0f, 0.0f, 0.0f); o glTranslatef(0.0f, 0.0f, -0.2f);
o glColor3f(0.0f, 0.8f, 0.0f); o glColor3f(0.5f, 0.5f, 0.5f);
o glutSolidTorus(0.2, 0.3, 10, 10);
o glPopMatrix();
o glPopMatrix(); //Cancel rotation
required for the tree to be upright
 Christmas Tree Example
o //Draw a flat star on top o //Draw some presents
o glPushMatrix(); o glPushMatrix();
o glTranslatef(0.0f, 3.0f, 0.0f); o glTranslatef(1.0f, 0.0f, 0.0f);
o glColor3f(1.0f, 1.0f, 0.0f); o glColor3f(0.5f, 0.0f, 0.8f);
o glBegin(GL_TRIANGLES); o glutSolidCube(0.5f);
o glVertex3f(-0.2f, -0.1f, 0.0f); o glPopMatrix();
o glVertex3f(0.2f, -0.1f, 0.0f); o glPushMatrix();
o glVertex3f(0.0f, 0.3f, 0.0f); o glTranslatef(0.8f, 0.0f, 0.8f);
o glVertex3f(-0.2f, 0.1f, 0.0f); o glColor3f(1.0f, 0.0f, 0.0f);
o glVertex3f(0.2f, 0.1f, 0.0f); o glutSolidCube(0.7f);
o glVertex3f(0.0f, -0.3f, 0.0f);
o glEnd();
o glPopMatrix();
o }
 Christmas Tree Example
o //Draw a bow on top of the large o glPopMatrix(); //Undo the
red present using 2 torus. transformations required for the
o glPushMatrix(); bow
o //Move to the top of the present o glPopMatrix();
o glTranslatef(-0.1f, 0.35f, -0.1f); o //Undo the transformations
o //Rotate about current x axis so required for the red present
the bow is flat. o glPushMatrix();
o glRotatef(90, 1.0, 0.0, 0.0); o glTranslatef(-0.8f, 0.0f, -0.8f);
o glColor3f(0.8f, 0.8f, 0.8f); o glColor3f(0.2f, 0.2f, 0.95f);
o glutSolidTorus(0.03, 0.12, 10, 10); o glutSolidCube(0.5f);
o glPushMatrix(); o glPopMatrix();
o //Centre of the other torus o glutSwapBuffers();
o glTranslatef(0.2f, 0.2f, 0.0f);
o glutSolidTorus(0.03, 0.12, 10, 10); o A timer is added to make the
o glPopMatrix(); whole scene rotate about the y
axis in world coordinates
Lecture Room Example
 Lecture Room Example
o /* This method simply creates o void createTable() { o // Create table top
a black pole by scaling and o // Create table legs o glPushMatrix();
translating a solid cube. */ o createPole(0.05,1.0,0.05); o glTranslatef(0.5, 1.025,
o glPushMatrix(); 0.5);
o void createPole(float xSize, o glTranslatef(0.0, 0.0, 0.95); o glColor3f(0.4509,0.2901,0
float ySize, float zSize) { .0705);
o glPushMatrix(); o createPole(0.05,1.0,0.05);
o glPopMatrix(); o glScalef(1.0, 0.05, 1.0);
o glTranslatef(xSize / 2, ySize / o glutSolidCube(1.0);
2, zSize / 2); o glPushMatrix();
o glTranslatef(0.95, 0.0, 0.95); o glPopMatrix();
o glColor3f(0.0,0.0,0.0); o } // end of createTable
o glScalef(xSize, ySize, zSize); o createPole(0.05,1.0,0.05);
o glutSolidCube(1.0); o glPopMatrix();
o glPopMatrix(); o glPushMatrix();
o } // end of createLeg o glTranslatef(0.95, 0.0, 0.0);
o createPole(0.05,1.0,0.05);
o /* The next method creates a o glPopMatrix();
table consisting of a table top
and four poles. The table top
is simply a scaled solid brown
cube. */
 Lecture Room Example
o /* This next method is used to
create a podium, using a long o /* This creates a row of tables. It creates
solid cube and a smaller and four tables, leave a gap, and then create
three more tables. Current position is
thinner cube at the top for a head where the table closest to the wall with the
board.*/ windows will appear. */
o void createPodium() { o void createRowOfTables(boolean fullRow) {
o // Create podium base o for(int i = 0; i < 4; i++) {
o glPushMatrix(); o glPushMatrix();
o glTranslatef(0.5, 0.75, 0.5); o glTranslatef(1.3 * i, 0.0, 0.0);
o createTable();
o glColor3f(0.4509,0.2901,0.0705); o glPopMatrix();
o glScalef(1.0, 1.5, 1.0); o } // end of for
o glutSolidCube(1.0); o if(fullRow) {
o glPopMatrix(); o glPushMatrix();
o // Create back board o glTranslatef(1.3 * 5, 0.0, 0.0);
o glPushMatrix(); o for(int i = 0; i < 3; i++) {
o glTranslatef(0.5, 1.51, 0.975); o glPushMatrix();
o glTranslatef(1.3 * i, 0.0, 0.0);
o glColor3f(0.4509,0.2901,0.0975); o createTable();
o glScalef(1.0, 0.2, 0.05); o glPopMatrix();
o glutSolidCube(1.0); o // end of for
o glPopMatrix(); o glPopMatrix();
o } // end of createPodium o } // end of if
o } // end of createRowOfTables()
 Lecture Room Example
o /* This method creates the walls o // Create back wall
of the room. Walls slightly vary in o glBegin(GL_QUADS);
colour just so that it gives an o glColor3f(1.0,1.0,1.0);
impression of shading helping in o glVertex3f(-5.5,-1.5,10.5);
distinguishing the difference o glVertex3f(-5.5, 1.5, 10.5);
between the walls. The current o glVertex3f(5.5, 1.5, 10.5);
position being the center of the
room. */ o glVertex3f(5.5, -1.5, 10.5);
o glEnd();
o void createWalls() {
o // Create right wall (wall with door)
o // Create front wall o glBegin(GL_QUADS);
o glBegin(GL_QUADS); o glColor3f(1.0,1.0,1.0);
o glColor3f(1.0,1.0,1.0); o glVertex3f(5.5, -1.5, -10.5);
o glVertex3f(-5.5,-1.5,-10.5); o glVertex3f(5.5, 1.5, -10.5);
o glVertex3f(-5.5, 1.5, -10.5); o glVertex3f(5.5, 1.5, 10.5);
o glVertex3f(5.5, 1.5, -10.5); o glVertex3f(5.5, -1.5, 10.5);
o glVertex3f(5.5, -1.5, -10.5); o glEnd();
o glEnd();
 Lecture Room Example
o // Create left wall (wall with windows) o glColor3f(0.90, 0.90, 0.90);
o glBegin(GL_QUADS); o glVertex3f(-3.5, 1.5, -10.5);
o glColor3f(1.0, 1.0, 1.0);
o glVertex3f(-3.5, 2.0, -10.5);
o glVertex3f(-5.5, -1.5, -10.5);
o glVertex3f(-5.5, 1.5, -10.5); o glVertex3f(-3.5, 2.0, 10.5);
o glVertex3f(-5.5, 1.5, 10.5); o glVertex3f(-3.5, 1.5, 10.5);
o glVertex3f(-5.5, -1.5, 10.5); o glEnd();
o glEnd(); o glBegin(GL_QUADS);
o // Create ceiling o glColor3f(0.90, 0.90, 0.90);
o glBegin(GL_QUADS);
o glVertex3f(-5.5, 2.0, -10.5);
o glColor3f(0.95, 0.95, 0.95);
o glVertex3f(-3.5, 1.5, -10.5); o glVertex3f(-5.5, 2.0, 10.5);
o glVertex3f(-3.5, 1.5, 10.5); o glVertex3f(-3.5, 2.0, 10.5);
o glVertex3f(5.5, 1.5, 10.5); o glVertex3f(-3.5, 2.0, -10.5);
o glVertex3f(5.5, 1.5, -10.5); o glEnd();
o glEnd(); o glBegin(GL_QUADS);
o // This creates the raised section in the
ceiling o glColor3f(1.0, 1.0, 1.0);
o glBegin(GL_QUADS); o glVertex3f(-5.5, 1.5, 10.5);
o glColor3f(1.0, 1.0, 1.0); o glVertex3f(-5.5, 2.0, 10.5);
o glVertex3f(-5.5, 1.5, -10.5); o glVertex3f(-3.5, 2.0, 10.5);
o glVertex3f(-5.5, 2.0, -10.5); o glVertex3f(-3.5, 1.5, 10.5);
o glVertex3f(-5.5, 2.0, 10.5);
o glEnd();
o glVertex3f(-5.5, 1.5, 10.5);
o glEnd(); o glBegin(GL_QUADS);
o glBegin(GL_QUADS); o
 Lecture Room Example
o glColor3f(1.0, 1.0, 1.0); o void createWindows() {
o glVertex3f(-5.5, 1.5, -10.5); o // Enable the texture
o glVertex3f(-5.5, 2.0, -10.5); o for(int i = 1; i < 5; i++) {
o glPushMatrix();
o glVertex3f(-3.5, 2.0, -10.5);
o glTranslatef(0.0, 0.75,3.9 * i);
o glVertex3f(-3.5, 1.5, -10.5); o glEnable(GL_TEXTURE_2D);
o glEnd(); o glBindTexture(GL_TEXTURE_2D,
o // Create floor lowWindows[i - 1]);
o glBegin(GL_QUADS); o createLowWindow();
o lDisable(GL_TEXTURE_2D);
o glColor3f(0.4549, 0.7333, 0.9843);
o glPopMatrix();
o glVertex3f(-5.5, -1.5, -10.5); o glPushMatrix();
o glVertex3f(-5.5, -1.5, 10.5); o glTranslatef(0.0, 2.022,3.9 * i);
o glVertex3f(5.5, -1.5, 10.5); o glEnable(GL_TEXTURE_2D);
o glVertex3f(5.5, -1.5, -10.5); o glBindTexture(GL_TEXTURE_2D,
o lEnd(); highWindows[i - 1]);
o createHighWindow();
o } // end of createWalls
o glDisable(GL_TEXTURE_2D);
o /* The next method adds four lower o glPopMatrix();
windows and 4 higher windows o glPopMatrix();
each with a different texture
mapping (using the textures stored o } // end of for
in highWindows and lowWindows. o } // end of createWindows
add all of the windows to the wall in
the room. The current position is
center of the room. */
 Lecture Room Example
o /* This creates a polygon with texture o /* This method will create the entire
mapping of an image of a projection. scene which will include the walls,
The current position is the center of windows, tables, chairs, podium,
where the image will appear. It will use doors, doorway and projection. */
the current projection (projectionNo) in
the projections array. The projection o void Display(void) {
goes in the x and y axis. */
o glLoadIdentity();
o void createProjection() { o glLightfv(GL_LIGHT1, GL_POSITION,
position);
o // Enable the texture o gluLookAt(x, y, z, x + lx,y + ly,z + lz,
o glEnable(GL_TEXTURE_2D); 0.0f,1.0f,0.0f);
o glBindTexture(GL_TEXTURE_ o glClear(GL_COLOR_BUFFER_BIT |
2D, projections[projectionNo]); GL_DEPTH_BUFFER_BIT);
o glBegin(GL_QUADS); o glEnable(GL_DEPTH_TEST)
o glTexCoord2f(0.0,0.0);
o glVertex3f(-1.568, -1.25, 0.0); o …
o glTexCoord2f(0.0,1.0); o …
o glVertex3f(-1.568, 1.25, 0.0); o } // end of Display
o glTexCoord2f(1.0,1.0);
o glVertex3f(1.568, 1.25, 0.0);
o glTexCoord2f(1.0,0.0);
o glVertex3f(1.568, -1.25, 0.0);
o glEnd();
o glDisable(GL_TEXTURE_2D);
o } // end of showProjection
Rubik’s Cube
 Modelling a Cube
{
1,2,3,4, // front face
y
Vertex array
5,6,7,8 // back face 7 -1,1,-1 Vi[0][1][2]:
3,2,7,6 // top face
1,3,5,4 // bottom face V1=(-1,-1,1)
3,6,5,4 // right face -1,1,1 2 6 1,1,-1 V2=(-1,1,1)
1,2,7,8 // left face V3=(1,1,1)
} V4=(1,-1,1)
o V5=(1,-1,-1)
cube[n].face[0].v[0].x=x-0.5; -1,-1,-1 8 3 1,1,1 V6=(1,1,-1)
cube[n].face[0].v[0].y=y-0.5; V7=(-1,1,-1)
cube[n].face[0].v[0].z=z-0.5;
V8=(-1,-1,-1)
cube[n].face[0].v[1].x=x+0.5;
cube[n].face[0].v[1].y=y-0.5; 1 -1,-1,1 5 1,-1,-1 x
cube[n].face[0].v[1].z=z-0.5;
cube[n].face[0].v[2].x=x+0.5;z
cube[n].face[0].v[2].y=y+0.5;
cube[n].face[0].v[2].z=z-0.5;
cube[n].face[0].v[3].x=x-0.5; 41,-1,1
cube[n].face[0].v[3].y=y+0.5;
cube[n].face[0].v[3].z=z-0.5;
 An Example Program
o #include <glut.h>
o GLfloat light_diffuse[] = {1.0, 0.0, 0.0, 1.0}; /* Red diffuse light. */
o GLfloat light_position[] = {1.0, 1.0, 1.0, 0.0}; /* Infinite light location. */

o GLfloat n[6][3] = { /* Normals for the 6 faces of a cube. */ y

o {-1.0, 0.0, 0.0}, {0.0, 1.0, 0.0}, {1.0, 0.0, 0.0}, -1,1,-1
o {0.0, -1.0, 0.0}, {0.0, 0.0, 1.0}, {0.0, 0.0, -1.0} }; 1,1,-1
-1,1,1 -1,-1,-1
o GLint faces[6][4] = { /* Vertex indices for the 6 faces of a cube. */
o {0, 1, 2, 3}, {3, 2, 6, 7}, {7, 6, 5, 4}, -1,-1,1 1,1,1 1,-1,-1
o {4, 5, 1, 0}, {5, 6, 2, 1}, {7, 4, 0, 3} };
x
o
o
GLfloat v[8][3]; /* array of 3D vectors (X,Y,Z values of vertices).*/
Void drawBox(void)
z 1,-1,1

o {
o int i; • faces[i][k] is index to kth vertex of the ith face
o for (i = 0; i < 6; i++) { • &v[faces[i][k]][0] is a pointer to kth vertex of ith face
o glBegin(GL_QUADS); • v[0][1][2]: x,y,z coordinates of a vertex
o glNormal3fv(&n[i][0]);
o glVertex3fv(&v[ faces[i][0] ] [0]);
o glVertex3fv(&v[ faces[i][1] ] [0]);
v[0][0] = v[1][0] = v[2][0] = v[3][0] = -1;
o glVertex3fv(&v[ faces[i][2] ] [0]); v[4][0] = v[5][0] = v[6][0] = v[7][0] = 1;
o glVertex3fv(&v[ faces[i][3] ] [0]); v[0][1] = v[1][1] = v[4][1] = v[5][1] = -1;
o glEnd(); v[2][1] = v[3][1] = v[6][1] = v[7][1] = 1;
o } v[0][2] = v[3][2] = v[4][2] = v[7][2] = 1;
o } v[1][2] = v[2][2] = v[5][2] = v[6][2] = -1;
 An Example Program
o Void display(void) o /* Use depth buffering for hidden surface
o { elimination. */
o glClear(GL_COLOR_BUFFER_BIT |
GL_DEPTH_BUFFER_BIT); o glEnable(GL_DEPTH_TEST);
o drawBox(); o /* Setup the view of the cube. */
o glutSwapBuffers(); o glMatrixMode(GL_PROJECTION);
o }
o Void init(void) /* field of view, aspect ratio, Z near, Z far */
o { o gluPerspective(40.0, 1.0, 1.0, 10.0);
o /* Setup cube vertex data. */
o v[0][0] = v[1][0] = v[2][0] = v[3][0] = -1; o glMatrixMode(GL_MODELVIEW);
o v[4][0] = v[5][0] = v[6][0] = v[7][0] = 1;
o v[0][1] = v[1][1] = v[4][1] = v[5][1] = -1; /*eye is (0,0,5), center is (0,0,0), up is Y direction */
o v[2][1] = v[3][1] = v[6][1] = v[7][1] = 1; gluLookAt(0.0, 0.0, 5.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.);
o v[0][2] = v[3][2] = v[4][2] = v[7][2] = 1;
o v[1][2] = v[2][2] = v[5][2] = v[6][2] = -1; /* Adjust cube position to be aesthetic angle*/
o /* Enable a single OpenGL light. */ o glTranslatef(0.0, 0.0, -1.0);
o glLightfv(GL_LIGHT0, GL_DIFFUSE, o glRotatef(60, 1.0, 0.0, 0.0);
light_diffuse); o glRotatef(-20, 0.0, 0.0, 1.0);
o glLightfv(GL_LIGHT0, }
GL_POSITION, light_position);
o glEnable(GL_LIGHT0);
o glEnable(GL_LIGHTING);
 Coursework

o Criteria
o Should not do exactly the same as the examples given.
Should be creative
o Submission deadline
o 4pm, Friday, 3rd December 2010
o Submission method
o Hardcopy of the report to the School Office on the
ground floor
o Email submission of a zip file, containing the source
code and softcopy of the report, to [email protected],
with the Subject: “G53GRA Coursework Submission”
 Practical This Week

o Getting started with OpenGL

o Try to run the Cube program
o Download BesierCurve.zip from course website
and try to understand and run the program