Homogeneous Coordinates en Slides

Geometric Fundamentals in Robotics
Homogeneous Coordinates
Basilio Bona
DAUIN-Politecnico di Torino
July 2009
Basilio Bona (DAUIN-Politecnico di Torino)
Rigid motions
July 2009
1 / 32
Introduction
Homogeneous coordinates are an augmented representation of points and
lines in Rn spaces, embedding them in Rn+1 , hence using n + 1 parameters.
This representation is useful in dealing with perspective and projective
transformation (computer graphics, etc.) and for rigid displacement
representation.
We start introducing the concept on 2D spaces.
Given a point p in R2 , represented as P = (p1 , p2 ), i.e., the vector

T
p = p1 p2 its homogeneous representation (using homogeneous
coordinates) is

T
= p1 p2 p3 ;
p

T
with 0 0 0 not allowed
The vector representation is obtained dividing the first n homogeneous

components by the (n + 1)-th, that is often called scale.
p1 = p1 /
p3 ;
p2 = p2 /
p3
Rigid motions
July 2009
2 / 32
A geometric interpretation of the homogeneous coordinates is given in the

following Figure.
Figure: Geometric interpretation of homogeneous coordinates.
Rigid motions
July 2009
3 / 32
We embed the plane : z = 1 in a 3D space and we consider the planar

coordinates as the intersection of a projection line that goes through the
origin of the RF {i, j, k}.
As z goes to we can represent the origin, while as z goes to 0, we can
represents points at infinity, along a well defined direction. The set of all
points at infinity represents the line at infinity.
Figure: Points at .
Rigid motions
July 2009
4 / 32
The same approach is valid for 3D homogeneous coordinates: given a

point p in R3 , represented as P = (p1 , p2 , p3 ), or else by a vector

T
p = p1 p2 p3
its homogeneous representation is

T
= p1 p2 p3 p4
p
Hence
p1 = p1 /
p4 ;
p2 = p2 /
p4 ;
p3 = p3 /
p4
In robotics it is common to set p4 = 1. We will adhere to this convention.
Rigid motions
July 2009
5 / 32
Some author treats differently the homogeneous representation of a point

(geometric vector vp ) by that of an oriented segment (physical vector vab ).
In the first case

v
vp = p
1
while in the second case

vab

v
v
v
= b a = ab
1
1
0
Not all textbooks adhere to this convention, since in this case the sum of
two geometric vectors produces

vp
vq
vp + vq
vp + vq =
+
=
1
1
2
This sum, once transformed back in 3D space, becomes not the usual
parallelogram sum of segments, but the midpoint of the parallelogram (see
also Affine Spaces).
Rigid motions
July 2009
6 / 32
Figure: Homogeneous sum of two geometric vectors.
Rigid motions
July 2009
7 / 32
Affine spaces
Briefly, affine spaces are spaces where the origin is not considered to be a
special point, and homogeneous coordinates are used to characterize the
affine elements (points and planes).
Figure: Linear and affine subspaces.
Rigid motions
July 2009
8 / 32
Rigid displacements and homogeneous coordinates

Using homogeneous representation, the generic displacement (R, t) is
given by the following homogeneous transformation matrix
"
#
R t
H H(R, t) = T
0 1
since
Hv =
"
0T
#
v
= Rv + t
1 1
t
Hence, the SE (3) group transformation g = (R, t) SE (3), is represented

by the 4 4 homogeneous matrix H through the use of homogeneous
coordinates:
a = Hab (Rab , tab )
pa = gab pb p
pb
Rigid motions
July 2009
9 / 32
Transformation composition gab gbc is obtained by the homogeneous

matrix product
Hab (Rab , tab )Hbc (Rbc , tbc )
i.e.,
"
#"
Rab tab Rbc
0T
0T
tbc
1
"
Rab Rbc
0T
Rab tbc + tab

1
Often it is convenient to write the time derivative of a homogeneous

vector as

p 1
p

= 2
p
p 3
0
This notation will prove useful when dealing with twists.
Rigid motions
July 2009
10 / 32
Introduction to projective/perspective geometry

are used for describing point in projective
Homogeneous coordinates p
spaces P n of dimension n.
are equal (
if there exist a real 6= 0 such
and b
Two points a
a b)
that
b
b
collinear points are equal
= b
a
this also means that vectors are equal up to a scale factor.
belong to the same line, i.e., to the same affine
and b
We say that a
subspace of Rn+1 .
Consider the planar (2D) projection through a pinhole camera of a 3D

T
point P, p = x y z . The ray of light intercepts the image plane in
the 2D image point Pi , whose homogeneous coordinates on the image
plane are

T

x
1 x T
x/f
i = y
p
pi =
=
y /f
f y
f
where f is the focal distance
Rigid motions
July 2009
11 / 32
Figure: Example of projection through a pinhole camera.

Rigid motions
July 2009
12 / 32
Considering the similar triangles in Figure, we have:

x
X
= ;
f
Z
y
Y
=
f
Z
If we set = Z /f , where f is known and the distance Z from of the

point p is unknown, we have
X = x,
Y = y ,
Z = f
Hence, if we compute the cartesian coordinates of pi on from

homogeneous coordinates, we have

fX
x
x
x/f
i = fY y y
pi =
p
y /f
Z
1
f

and every 3D point with homogeneous coordinates x
projected onto the same point pi of the image plane .
Notice that the above transformation is a nonlinear one.

Rigid motions
T
is
July 2009
13 / 32
The nonlinear mapping R3 R2 can be expressed as a (linear) matrix

transformation between the homogeneous spaces R4 R3 as

X

fX
x
f 0 0 0
p
Y
fY y = 0 f 0 0
Z pi = P 1
Z
1
0 0 1 0
{z
} 1
|
P
Nonsingular (i.e., invertible) mappings between projective spaces are

described by products of full rank square matrices T by homogeneous
a = Tab p
b
coordinates, p
An invertible mapping from P n onto itself preserves collinearities, i.e., if

two points belong to a line, also the transformed points will belong to a
line.
Affine transformations preserve not only collinearities, but also parallelism
between lines and ratios of distances.
Similarity transformations preserves also angles, but not scales, while
Euclidean (rigid) transformations preserve also scales
Rigid motions
July 2009
14 / 32
Figure: Examples of different mappings.

Rigid motions
July 2009
15 / 32
Hyperplanes
A hyperplane in Rn is defined by the equation between a set of points pi
and a set of (real) coefficients i
1 p1 + 2 p2 + + n pn + n+1 = 0
Using the homogeneous coordinates we obtain an homogeneous linear
equation in P n
T p
=0
where
1
.
= .. ;

n
n+1

p1
..
.
=
p
p
n
1
The two homogeneous vectors can exchange their roles; we say that they
are dual.
Rigid motions
July 2009
16 / 32
Remember that, given a vector p V(F), the set of all linear

transformations f (p) : V 7 R1 form another vector space V , and
f V are called dual vectors or 1forms.
For example, the total time derivative of a scalar function f (p) on R3 is
df (p(t))
= p (p)p(t)
dt
Another dual vector is the divergence
is a dual vector wrt p.

p1
T .
..
p=
p1
pn
pn
or the Laplacian (gradient of the divergence)
f = f =
p1
f
p1
T
..
.
pn
f
pn
Rigid motions
July 2009
17 / 32
Duality in projective spaces
The space of all hyperplanes in Rn defines another perspective space,

called the dual of the original P n
Duality principle: for all projective results established using points and
planes, a symmetrical result holds interchanging the role of planes and
points.
Example:
two points define a line
two lines define a point
Notice that the projective plane P 2 whose point are defined by

homogeneous coordinates, has more point that the usual R2 plane, since it
includes also points at .
Rigid motions
July 2009
18 / 32
A collineation given by a full rank matrix T transforms points as

a = Tab p
b
p
T
b = 0 is transformed as
A hyperplane b p
T
T
b = 0 a p
a = 0
bp
hence
yielding
T
T
T
b b = a Tab
a Tab p
a = T T T
b
ab
T
Matrix T
ab is called the dual of Tab
Comparing this last equation with the first one, we conclude that points
and lines can be transformed, but using different (dual) collineation
T
matrices Tab and T
ab .
Rigid motions
July 2009
19 / 32
Points and lines

We restrict our analysis on the 2D projective space P 2 ; for the sake of
is
clarity, we drop the sign on top of the homogeneous vectors, i.e., a
now indicated as a.

A line = 1 2 3 can be characterized by three parameters:
1
2
3
the slope 1 /2
the x-intercept 3 /1
the y -intercept 3 /2
Points p and lines are related by

T p = 0
in the sense that a point is on a line or a line include a point iff the
previous relation holds.
Rigid motions
July 2009
20 / 32
Duality
A line passing for two points p1 and p2 is defined as
= p1 p2 = S(p1 )p2
A point p belonging to two lines (intercept) 1 and 2 is defined as
p = 1 2 = S(1 )2
Points at infinity or ideal points have the following representation

T
p = p1 p2 0
All ideal points lie on the ideal line, represented by

T
= 0 0 1
Rigid motions
July 2009
21 / 32
Three points p1 , p2 , p3 lie on the same line (they are collinear) iff

det p1 p2 p3 = 0
Since the duality principle holds, three lines 1 , 2 , 3 meet at the same
point (they are concurrent) iff

det 1 2 3 = 0
Rigid motions
July 2009
22 / 32
Parallelism
Two lines
m1
m = m2 ;
m3

n1
n = n2
n3
are parallel when the intersection point is at infinity, i.e.,

p1
p = m n = p2
0
This reduces to the conditions

m1
n1
m1 n1
=0
=
or det
m2 n2
m2
n2

p
The direction k 1 points toward the point at infinity.
p2
Rigid motions
July 2009
23 / 32
Summary

T
p = p1 p2 p3
line
pT = 0
incidence
line
= p1 p2
point
p = 1 2
collinearity

det p1 p2 p3 = 0
collinearity

det 1 2 3 = 0
point
incidence
point
p1 p2 0
T
line
Rigid motions

T
= 1 2 3
T p = 0
0 0 3
T
July 2009
24 / 32
Transformations in P 2
The generic projective transformation (also called homographies) in P 2 is
represented by a non singular 3 3 matrix (acting on homogeneous
vectors), whose elements are
t11 t12 t13

T = t21 t22 t23
t31 t32 t33
Since the transformations are equivalent up to a scale factor, the number

of dof in T is 8.
According to structural constraint, we have a hierarchy of transformations,

from the most general to the least one; in particular
Transformations in P 2
(most general) projective affine similarity Euclidean (rigid)
Rigid motions
July 2009
25 / 32
The affine transformation must preserve the ideal line and the ideal points
(as well as parallelism); hence, for any arbitrary scaling factor

p1
p1
p2 = T p2
0
0
that implies
t11 t12 t13

T = t21 t22 t23
0
0 t33
Rigid motions
dof=6
July 2009
26 / 32
The similarity transformation must preserve angles and ratios of lengths;

that implies
cos sin t13

T = sin cos t23 dof=3
0
0
t33
with t33 6= 1.
The effects of the last column in T is to produce a shear i.e., a scale

variation that is non uniform in all directions.
Rigid motions
July 2009
27 / 32
The Euclidean (rigid) transformation must preserve also lengths; that

implies
cos sin t13

T = sin cos t23 dof=3
0
0
1
Rigid motions
July 2009
28 / 32
Historical notes
Homogeneous coordinates were introduced in 1827 by M
obius in the
context of barycentric coordinates. Given a triangle, the barycentric
coordinates of a point P are the weights required at the triangle vertices
such that P become the center of gravity of the triangle.

T
The point is computed as p = wa x wb y wc z , with
wa + wb + wc = 1
Figure: Example of barycentric coordinates.

Rigid motions
July 2009
29 / 32
From homogeneous coordinates to quaternions

There exist a link between homogeneous coordinates and quaternions that
goes through the M
obius transformation.
Given two complex numbers x = x1 + jx2 , y = y1 + jy2 C, the ordered
pair of complex numbers

z = x
T
C2 ,

T
z 6= 0 0
are the homogeneous coordinates of z = z1 + jz2

z=
x
x1 + jx2
= z1 + jz2 =
y
y1 + jy2
To every pair of complex numbers (x, y ) a unique complex numbers z

corresponds, while for every complex number z there is an infinite number
of complex pairs (projective transformation).
Rigid motions
July 2009
30 / 32
Transformation between homogeneous representations

x
u
u
a b x
ax + by
z =
=
7 w
where
=
=
y
v
v
c d y
cx + dy
where a, b, c, d are complex numbers.
We can compute the transformed complex number as
w=
a xy + b
u
ax + by
az + b
=
= x
=
v
cx + dy
cy +d
cz + d
This is nothing else that the M

obius transformation.
Mobius transformation
w=
az + b
;
cz + d
a, b, c, d, w , z C
Rigid motions
July 2009
31 / 32
If we take a particular form of the M

obius transformation, due to Gauss:
w=
az + b
b z + a
we know that every rotation of the sphere can be expressed by a complex

function as the one above, and we notice that it can be represented by the
matrix of its coefficients

a
b
+ j + j
=
b a
+ j j

1 0
j 0
0 1
0 j
=
+
+
+
0 1
0 j
1 0
j 0
= 1 + i + j + k
= a quaternion
For further details see: T. Needham, Visual Complex Analysis, OUP, 1997.
Rigid motions
July 2009
32 / 32

Homogeneous Coordinates en Slides

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Homogeneous Coordinates en Slides

Uploaded by

Copyright:

Available Formats

Geometric Fundamentals in Robotics

Basilio Bona (DAUIN-Politecnico di Torino)

The vector representation is obtained dividing the first n homogeneous

A geometric interpretation of the homogeneous coordinates is given in the

Figure: Geometric interpretation of homogeneous coordinates.

Basilio Bona (DAUIN-Politecnico di Torino)

We embed the plane : z = 1 in a 3D space and we consider the planar

Basilio Bona (DAUIN-Politecnico di Torino)

The same approach is valid for 3D homogeneous coordinates: given a

its homogeneous representation is

In robotics it is common to set p4 = 1. We will adhere to this convention.

Basilio Bona (DAUIN-Politecnico di Torino)

Some author treats differently the homogeneous representation of a point

while in the second case

Figure: Homogeneous sum of two geometric vectors.

Basilio Bona (DAUIN-Politecnico di Torino)

Figure: Linear and affine subspaces.

Basilio Bona (DAUIN-Politecnico di Torino)

Rigid displacements and homogeneous coordinates

Hence, the SE (3) group transformation g = (R, t) SE (3), is represented

Basilio Bona (DAUIN-Politecnico di Torino)

Transformation composition gab gbc is obtained by the homogeneous

Rab tbc + tab

Often it is convenient to write the time derivative of a homogeneous

This notation will prove useful when dealing with twists.

Basilio Bona (DAUIN-Politecnico di Torino)

Introduction to projective/perspective geometry

Basilio Bona (DAUIN-Politecnico di Torino)

Figure: Example of projection through a pinhole camera.

Considering the similar triangles in Figure, we have:

If we set = Z /f , where f is known and the distance Z from of the

Hence, if we compute the cartesian coordinates of pi on from

Notice that the above transformation is a nonlinear one.

The nonlinear mapping R3 R2 can be expressed as a (linear) matrix

Nonsingular (i.e., invertible) mappings between projective spaces are

An invertible mapping from P n onto itself preserves collinearities, i.e., if

Figure: Examples of different mappings.

Remember that, given a vector p V(F), the set of all linear

or the Laplacian (gradient of the divergence)

Basilio Bona (DAUIN-Politecnico di Torino)

Duality in projective spaces

The space of all hyperplanes in Rn defines another perspective space,

two lines define a point

Notice that the projective plane P 2 whose point are defined by

Basilio Bona (DAUIN-Politecnico di Torino)

A collineation given by a full rank matrix T transforms points as

Points and lines

Points p and lines are related by

Basilio Bona (DAUIN-Politecnico di Torino)

All ideal points lie on the ideal line, represented by

Basilio Bona (DAUIN-Politecnico di Torino)

are parallel when the intersection point is at infinity, i.e.,

Basilio Bona (DAUIN-Politecnico di Torino)

t11 t12 t13

Since the transformations are equivalent up to a scale factor, the number

According to structural constraint, we have a hierarchy of transformations,

t11 t12 t13

Basilio Bona (DAUIN-Politecnico di Torino)

The similarity transformation must preserve angles and ratios of lengths;

cos sin t13

The effects of the last column in T is to produce a shear i.e., a scale

Basilio Bona (DAUIN-Politecnico di Torino)

The Euclidean (rigid) transformation must preserve also lengths; that