Vision based Real-time Modeling of Dynamic Unstructured
Environments in Driving Scenarios
Andrei Vatavu and Sergiu Nedevschi
Computer Science Department, Technical University of Cluj-Napoca, 26-28 G. Baritiu Street, Cluj-Napoca, Romania
Keywords: Environment Representation, Polyline, Object Contour, Iterative Closest Point, Driving Assistance,
Stereo-vision, Object Delimiters, Motion Detection.
Abstract: The detection of moving traffic participants is an essential intermediate step for higher level driving
technology tasks. Regardless of the type of used sensors, dynamic environment modeling becomes even
more difficult when the surrounding world is unstructured and heterogeneous. In such complex
environments the representation system can be affected by noisy measurements, occlusions, wrong data
association or unpredictable nature of the traffic participants. We propose a solution of representing the
dynamic environment in real-time by using the pairwise alignment of free-form models and considering the
advantages provided by a dense stereovision system. Instead of registering the whole 3D point cloud, our
method is based on extracting and registering a more compact model of the environment taking into
consideration the most visible object cells from the ego car. The proposed method is based on information
provided by a Digital Elevation-Map, but can be easily adapted for other types of intermediate
representations.
1 INTRODUCTION
In the context of Advanced Driver Assistance
Systems, modeling static and dynamic entities of the
environment is a key problem. The detection of
moving traffic participants is an essential
intermediate step for higher level driving technology
tasks such as collision warning and avoidance, path
planning or parking assistance. The problem of
dynamic environment representation becomes even
more difficult when the surrounding world is
unstructured and heterogeneous, including the cases
of crowded urban centers, traffic intersections or off-
road scenarios. The representation component may
be influenced by several factors: noisy
measurements, occlusions, wrong data association or
unpredictable nature of the traffic participants. In
such complex environments, a driver assistance
system should be able to detect other moving traffic
entities in real-time and at a high accuracy.
Usually, the classic approaches of dynamic
obstacles detection and tracking consist in extracting
a set of features from the scene and estimating the
motion from their displacement. Current solutions
can directly use 3D points (Franke, 2005), or they
can track high level attributes such as 2D boxes or
3D cuboids (Danescu, 2007), stixels (Pfeiffer, 2010),
free-form polygonal models (Wang, 2007), object
contours (Prakash, 2007; Yokoyama, 2005) etc.
The dynamic obstacle modeling solutions can be
classified by the nature of used sensors. The most
common used sensors are vision based (Danescu,
2007), laser (Thomas, 2010; Madhavan, 2002),
sonar (Fairfield, 2007) or radar. The motion
estimation techniques are also distinguished by the
level at which the dynamic features detection is
applied. Some of the existing methods rely on
computing motion before generating a model
(Danescu, 2012; Hess, 2008), while other methods
are based on extracting some attributes and
subsequently estimating their dynamic parameters
(Danescu, 2007; Wang, 2007 and Prakash, 2007).
Many of dynamic object detection solutions use
intermediate representations as primary information.
A common practice is mapping 3D information into
occupancy grids (Danescu, 2012), digital elevation
maps (Danescu, 2009) or octrees (Fairfield, 2007).
The data association and identifying correct
correspondences steps play an important role in
estimating the motion of the traffic entities. One of
the widely used methods for model fitting in the
presence of many data outliers is the RANSAC
141
Vatavu A. and Nedevschi S..
Vision based Real-time Modeling of Dynamic Unstructured Environments in Driving Scenarios.
DOI: 10.5220/0004045401410149
In Proceedings of the 9th International Conference on Informatics in Control, Automation and Robotics (ICINCO-2012), pages 141-149
ISBN: 978-989-8565-22-8
Copyright
c
2012 SCITEPRESS (Science and Technology Publications, Lda.)
algorithm (Fischler, 1981). However, its accuracy
depends directly on the number of used samples.
This may lead to a high computational cost.
Direct matching solutions such as Iterative
Closest Point (ICP) (Besl and McKay, 1992)
algorithm are most common for vehicle localization
and mapping (Wang, 2007). In (Rusinkiewicz, 2001)
the convergence performance for several ICP
variants is compared. An optimized ICP method that
uses a constant time variant for finding the
correspondences is presented. In (Wang, 2007) a
moving objects map is segmented by assuming that
dynamic parts do not fulfill the constraints of the
SLAM. However, the most of scan matching
methods do not take into consideration the ego-
motion parameters. The data association of objects
in subsequent scans is hard to be achieved when the
traffic participants or the ego vehicle moves at high
speeds or when the measurement uncertainties are
not taken into account.
We propose a solution of representing the
dynamic environment in real-time by using the
pairwise alignment of free-form delimiters and
considering the advantages provided by a
stereovision system, by inheriting the object
information from the intermediate representation.
Instead of registering the whole 3D point cloud, our
method is based on extracting the most visible object
cells from the ego car and using them as input data
for the alignment process. We propose an extension
of the classical ICP algorithm by applying a set of
improvement heuristics:
The data association is one of the problems of
the classical scan matching techniques. It’s
hard to estimate the correspondent models
from previous scans only based on the
proximity criterion. In our case we introduce a
pre-processing step. First, we find the
correspondence pairs between the model set
(contour extracted in previous frame) and the
measurement set (current frame results) by
finding similarities between object blobs and
passing this information at the contour level.
Then, a list of associated contour candidates is
generated and is used as the input for the next
steps of the alignment;
For the registration process we use free-form
polygonal models that minimize the erroneous
results caused by occlusions, or by stereo
reconstruction errors. The main idea is that we
are taking into account only the most visible
points from the ego-vehicle by performing a
radial scanning of the environment (Vatavu,
2009);
The previously extracted speeds are used as
the initial guess for the ICP algorithm;
In order to filter the alignment outliers, a
rejection metric that includes stereo
uncertainties is proposed;
Our method is based on information provided by a
Digital Elevation-Map, but can be easily adapted for
other types of intermediate representations.
The remaining of the paper is structured as
follows: Section 2 introduces the architecture of the
proposed dynamic environment representation.
Section 3 presents the pre-processing module with a
group of necessary tasks for extracting object
dynamic properties. In section 4, the main steps of
the motion estimation component are detailed. The
last two sections show the experimental results and
conclusion about this contribution.
2 SYSTEM ARCHITECTURE
The dynamic environment representation method
has been developed and adapted for crowded
environments such as urban city traffic scenes. In
this paper we extend our previous Dense Stereo-
Based Object Recognition System (DESBOR)
(Nedevschi, 2007). The system architecture could be
divided in four main blocks: data acquisition and 3D
reconstruction, intermediate representation, pre-
processing, and motion estimation.
Data Acquisition and 3D Reconstruction is the
first level of the processing flow. At this stage the
images are acquired from the two cameras, then the
3D reconstruction is performed using a specialized
TYZX (Woodill, 2004) board. The resulted point
cloud is used as the input information for computing
the Digital Elevation Map.
Intermediate Representation: the raw dense
stereo information is mapped into a Digital
Elevation Map. The resulted intermediate
representation contains three types of cells: road,
traffic isle and object. The cells are labeled based on
their height information. More details about the
Elevation Map are presented in (Oniga, 2010).
Pre-processing: The pre-processing level groups
a set of basic tasks that are performed prior the ICP
algorithm. At this phase, the object contours are
extracted by radial scanning of the Elevation Map.
For the delimiters extraction we use the Border
Scanner algorithm previously developed by us
(Vatavu, 2009). We apply the ego-motion
compensation for the Elevation Map and contours
that are extracted in previous frame, assuming that
we know the odometry information. The ego-vehicle
ICINCO2012-9thInternationalConferenceonInformaticsinControl,AutomationandRobotics
142
motion is compensated in order to separate its speed
from the independent motion of the objects in the
traffic scene. Another pre-processing task is to
associate the polygonal models. The data association
is achieved by using the maximum overlapping
score of the Elevation Map blobs. Considering that
each polygonal model inherits the blob type, it also
inherits the blob association information.
Figure 1: System Architecture.
Motion Estimation: As the result of the pre-
processing level, a list of candidates is provided for
the ICP module. Each candidate represents a pair of
associated contours in the previous stage. For each
candidate, a rotation and a translation is estimated by
the ICP algorithm. Then the computed motion
information is associated to the static polygonal
models. A dynamic polyline map is generated as the
result. Each polyline element is characterized by a
set of vertices describing the polygon, position,
height, type (traffic isle, obstacle), orientation and
magnitude.
In our case the two cameras are placed on a
moving vehicle. We use a coordinate system where
the z axis points toward the direction of the ego-
vehicle, and the x axis is oriented to the right. The
origin of the coordinate system is situated in front of
the car (Figure 3).
3 PRE-PROCESSING LEVEL
The pre-processing stage consists in performing
necessary tasks prior the motion estimation. First,
extracting a sufficiently generic model is needed.
The extracted model should allow us the creation of
fast subsequent algorithms and as well it should
minimize the representation errors caused by noisy
3D reconstruction or by occlusions.
Figure 2: a) An urban traffic scene. b) The Elevation Map
projected on the left camera image. d) The top view of the
Elevation Map. The Elevation Map cells are classified
(blue – road, yellow – traffic isle, red – obstacles). c) A
compact representation of the environment. The extracted
polygonal models are considered static. The extracted
delimiters inherit the object information from the
Elevation Map (green – obstacles, yellow – traffic isles).
A second task is to separate the ego-vehicle
speed from the independent motion of the other
objects in the traffic scene. This is achieved by
compensating the ego motion.
And finally, elevation map blob is labeled and is
used in data association. As the result a list of pairs
of contours is extracted and is provided subsequently
to the ICP step. Thus, unlike the other classical
methods that involve aligning the whole local maps
at once, and then segmenting the dynamic obstacles
VisionbasedReal-timeModelingofDynamicUnstructuredEnvironmentsinDrivingScenarios
143
from the static ones, we first associate the obstacles
at the blob level and then apply the ICP for each
associated candidate.
Figure 3: Coordinate System.
3.1 Polyline-based Environment
Perception
For the polyline based object representation we use
the Border Scanner algorithm described by us in
(Vatavu, 2009). The main idea is that we are taking
into account only the most visible points from the
ego car and extract object delimiters by radial
scanning of the Elevation Map. Our method is
similar to a Ray-Casting approach. The proposed
method consists in determining the first occupied
point intersected by a virtual ray which extends from
the ego-car position. The scanning axis moves in the
radial direction, having a fixed center at the ego-
vehicle position (the coordinate system origins). At
each step we try to find the nearest visible point
from the Ego Car situated on the scanning axis. In
this way, all subsequent cells P
i
are accumulated into
a Contour List C, by moving the scanning axis in the
radial direction:
),...,{
21 n
PPPC =
(1)
For each object O
i
described by a contour C
i
we
apply a polygonal approximation of C
i
by using a
split-and-merge technique described in (Douglas and
Peuker, 1973). The extracted polygon is used to
build a compact 3D model based on the polyline set
of vertices as well as on the object height. A polyline
based representation is described in Figure 2.d.
3.2 Ego-motion Compensation
Before estimating the motion of the traffic entities,
the movement of the ego vehicle must also be taken
into consideration. In order to compensate the ego
motion in the successive frames, for each given
point P
t-1
(x
t-1
, y
t-1
, z
t-1
) in the previous frame, the
corresponding coordinates P
t
(x
t
, y
t
, z
t
) in the current
frame are computed by applying the following
transformation:
()
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
+
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
=
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
−
−
−
zt
t
t
y
t
t
t
tz
y
x
R
z
y
x
0
0
1
1
1
ψ
(2)
Where
(
)
ψ
y
R
is the rotation matrix around the
Y axis with a given angle
ψ
, and t
z
is the translation
on the Z axis. The rotation and the translation
parameters are provided by the ego-car odometry. It
is considered that the translations on the X and Y
axis are zero.
3.3 Data Association
This stage consists in finding the corresponding
contours that identify a single object in consecutive
frames. As each extracted contour describes an
Elevation Map blob, finding the associated contour
pairs is reduced to find a similarity between the
object blobs.
For each object P
i
from the previous frame and
for each object C
j
from the current frame we
calculate an overlapping score A
ij
. The results are
stored into a score matrix A={A
ij
}. Candidates with
the highest score are taken into account in
determining the associations between the two set of
objects P and C.
Figure 4: The association between two set of blobs in the
consecutive frames and the resulting set of associated
pairs.
However the association problem may lead only
ICINCO2012-9thInternationalConferenceonInformaticsinControl,AutomationandRobotics
144
to partial results in the cases when larger objects
from the previous frame are split into smaller blobs
in the current frame and vice versa. In order to find
all possible pairs of candidates we perform two
types of associations: a direct association (forward
association) finding best overlapping candidates in
the current frame for all blobs in the previous frame,
and a reverse association (backward association) that
finds best overlapped objects in the previous frame
for all objects from the current frame. The final list
of candidates includes all distinct pairs associated in
the two steps.
4 MOTION ESTIMATION
The object motion estimation module receives as
input a list of associated contour pairs. For each
distinct pair we compute correspondences between
the two contours and estimate a rotation and a
translation which minimize the alignment error. For
the contour pairwise registration we use the Iterative
Closest Point (ICP) method. The ICP algorithm was
proposed by Besl and McKay (Besl and McKay,
1992) and represents a common solution especially
for scan-matching techniques, but the idea could be
adapted for any kind of models.
For each contour pair that identifies the same
object in the consecutive frames we define two set of
points: a model set P={p
1
,p
2
, ..., p
M
} that describes
the object contour in the previous frame, and a data
set Q={q
1
,q
2
, ..., q
K
} that describes the object
contour in the current frame. For each point q
j
from
Q the corresponding closest point p
i
from P is found.
We want to find an optimal rotation R and
translation T that minimize the alignment error. The
objective function is defined:
∑
=
−+=
N
i
ii
qTRpTRE
1
2
),(
(3)
where p
i
and q
i
are the corresponding point pairs
of the two sets and N is the total number of
correspondences.
The proposed alignment method is described by
the following main steps:
1.Matching – for each point from data set, the
closest point from the model set is found. A list of
correspondent pairs is generated.
2.Outliers Rejection – Rejecting the outliers that
could introduce a bias in the estimation of
translation and rotation.
3.Error Minimization – estimating new
transformation parameters R and T for the next
iteration.
4.Updating – having the new R and T, a new target
set is computed by applying the new transformation
to the model set. A global transformation M
g
is
updated with the new R and T values.
5.Testing the Convergence – compute the average
point-to-point distance between the measurement set
and transformed model set. Then test if the
algorithm has been converged to a desired result. If
the error is greater than a given threshold, the
process continues with a new iteration. The
algorithm stops when the computed error is below
the selected error threshold or when a maximum
number of iterations have been achieved.
Next we will detail each of these steps.
4.1 Matching
At this stage, for each point q
i
from Q we want to
find the closest point from the model set P:
),(min),(
}..1{
ji
Nj
i
pqdPqd
p
∈
=
(4)
Usually this task is the most computationally
extensive in the ICP algorithm. The classical brute
force search approach has a complexity of
)(
pq
NNO
⋅
, with N
p
being the number of points
in P and N
q
– the number of points in Q. In order to
Figure 5: Distance Transforms and Corresponding Masks
are computed for dynamic obstacles (left side), and for
static obstacles (right side). Data contours (gray color) and
model contours (white color) are superimposed on the
Distance Transform image. Each contour point in the
correspondence mask is labeled with a unique color. The
colors in the corresponding mask identify uniquely the
closest contour point (having the same color).
VisionbasedReal-timeModelingofDynamicUnstructuredEnvironmentsinDrivingScenarios
145
reduce the complexity to
)log(
pq
NNO ⋅
many
solutions employ a KD-Tree (Bentley, 1975) data
structure. In our case, for finding closest points
problem, we use a modified version of Chamfer
based Distance Transform (Borgefors, 1984).
A distance transform represents a map that has
the property that each map cell has a value
proportional to the nearest obstacle point.
In our case, for each separate model contour we
define a region of interest and compute the distance
transform. The difference of our solution is that we
use two maps: a distance map that store the
minimum distances to the closest points, and a
correspondence map, storing the positions of the
closest points (Figure 5). The correspondences from
the model set are identified by superimposing the
data contour on the two masks.
4.2 Outliers Rejection
The purpose of this stage is to filter erroneous
correspondences that could influence the alignment
process. We use two types of rejection strategies:
rejection of pairs whose point-to-point distance is
greater than a given threshold, and eliminating the
points where the overlap between the two contours
is not complete.
4.2.1 Distance based Rejection
The classical strategy consists in rejection of pairs
whose point-to-point distance is larger than a given
threshold D
t
:
tji
Dpqd >),(
(5)
Because the stereo reconstruction error generally
increases with the square of the z distance, the
stereo-system uncertainties must be taken into
account. As suggested by (Danescu, 2012), if we
assume that the stereo-vision system is rectified,
then the z error is given by the following relation:
fb
z
d
z
⋅
⋅
=
σ
σ
2
(6)
Where z is the depth distance, b is the stereo
system baseline; f is the focal length and
d
σ
denotes
the disparity error.
Thus, for each corresponding pair of points (p
i
,q
i
)
from the two sets, the rejection is made if:
ztji
Dpqd
σ
+>),(
(7)
This would mean that the rejecting threshold is
increased at once with the z distance.
4.2.2 Boundary based Rejection
The second type of rejecting is filtering the point
correspondences caused by incomplete overlap
between contours. Usually, these situations appear
when one of the two contours is incompletely
extracted due to occlusions, and may lead to
incorrect alignments.
A possible solution is to identify the subsets of
points from Q that have the same correspondent
point p
j
in P, and keeping only the pair with the
minimum distance (Figure 6).
Figure 6: Rejecting the contour boundary.
4.3 Error Minimization
In this step we determine the optimal rotation R and
translation T by minimizing the objective function
defined by Equation (3).
The rotation matrix around the Y axis is linearized,
approximating
α
cos
by 1 and
α
sin
by
α
:
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
−
≈
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
−
=
10
010
01
cos0sin
010
sin0cos
)(
α
α
αα
αα
α
y
R
(8)
The translation vector is defined as:
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
=
z
y
x
t
t
t
T
(9)
We can rewrite the Equation (3) as:
∑
=
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
−
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
+
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎣
⎡
−
=
N
i
zi
yi
xi
z
y
x
zi
yi
xi
q
q
q
t
t
t
p
p
p
TRE
1
2
,
,
,
,
,
,
10
010
01
),(
α
α
(10)
The
),( TRE
is minimized with respect to
α
, t
x
, t
y
,
and t
z
by setting the partial derivatives to zero:
ICINCO2012-9thInternationalConferenceonInformaticsinControl,AutomationandRobotics
146
()
()
()
⎪
⎪
⎪
⎪
⎪
⎩
⎪
⎪
⎪
⎪
⎪
⎨
⎧
=−−+=
∂
∂
=−+=
∂
∂
=−++=
∂
∂
=
⎥
⎥
⎦
⎤
⎢
⎢
⎣
⎡
+−−
++
=
∂
∂
∑
∑
∑
∑
=
=
=
=
02
),(
02
),(
02
),(
0
)(
2
),(
1
,,,,
1
,,,
1
,,,,
1
,,,,,,
,,
2
,
2
,
N
i
zixizizi
z
N
i
yiyiyi
y
N
i
xizixixi
x
N
i
xizizixixizi
zixizixi
qppt
t
TRE
qpt
t
TRE
qppt
t
TRE
pqpqpt
ptpp
TRE
α
α
α
α
(11)
Therefore we can obtain the unknown
coefficients:
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎩
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎨
⎧
⎟
⎠
⎞
⎜
⎝
⎛
+−=
⎟
⎠
⎞
⎜
⎝
⎛
−=
⎟
⎠
⎞
⎜
⎝
⎛
−−=
⎟
⎟
⎠
⎞
⎥
⎦
⎤
⎢
⎣
⎡
−+
⎜
⎝
⎛
+−⋅
⋅
⎟
⎠
⎞
⎜
⎝
⎛
−
⎟
⎠
⎞
⎜
⎝
⎛
−+
=
∑∑∑
∑∑
∑∑∑
∑∑
∑∑∑∑
∑∑∑
===
==
===
==
====
===
N
i
xi
N
i
zi
N
i
ziz
N
i
yi
N
i
yiy
N
i
zi
N
i
xi
N
i
xix
N
i
xizi
N
i
zixi
N
i
N
i
xizi
N
i
N
i
zixi
N
i
N
i
zi
N
i
xizixi
ppq
N
t
pq
N
t
ppq
N
t
pqpqN
qpqp
ppppN
1
,
1
,
1
,
1
,
1
,
1
,
1
,
1
,
1
,,
1
,,
11
,,
11
,,
1
2
1
,
2
1
,
2
,
2
,
1
1
1
)()(
)(
1
α
α
α
(12)
4.4 Updating
Assuming that we have estimated new R and T
parameters in the previous step, a new target set is
computed by applying the new transformation to the
model set.
Having the rigid body transformation matrix M:
⎥
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎢
⎣
⎡
−
=
⎥
⎦
⎤
⎢
⎣
⎡
=
1000
10
010
01
10
z
y
x
t
t
t
TR
M
α
α
(13)
Each point p
i
from the model set P is transformed
according to the following relation:
⎥
⎥
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎢
⎢
⎣
⎡
⎥
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎢
⎣
⎡
−
=
⎥
⎥
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎢
⎢
⎣
⎡
+
+
+
1
1000
10
010
01
1
,
,
,
1
,
1
,
1
,
k
zi
k
yi
k
xi
z
y
x
k
zi
k
yi
k
xi
p
p
p
t
t
t
p
p
p
α
α
(14)
Finally, a global transformation M
G
is updated:
MMM
GG
⋅
=
(15)
4.5 Testing the Convergence
The error metric is estimated by computing the
average Euclidean distance (AED) of every
corresponding pair of data set Q and transformed
model set.
∑
=
−=
N
i
ii
qp
N
Err
1
1
(16)
If the error is greater than a given threshold, the
process continues with a new iteration. The
algorithm stops when the computed error is below
the selected error threshold or when a maximum
number of iterations have been achieved.
5 EXPERIMENTAL RESULTS
The proposed dynamic environment representation
method has been tested in different traffic situations.
For our experiment we used a 2.66GHz Intel Core 2
Duo Computer with 2GB of RAM. Figure 7 presents
some qualitative results obtained in a dynamic urban
traffic scenario. In figure 7.b the model delimiter
that was extracted in previous frame is colored with
yellow, while the data contour (extracted in the
current frame) is drawn with blue. The result of the
alignment is illustrated with red color. It can be
observer that in the case of the incoming vehicle, as
well as for the lateral static vehicles, the aligned
model is superimposed almost perfectly on the data
set. In the Figure 7.c, the virtual view of the scene is
shown. The static obstacles are represented with
green delimiters while the dynamic obstacles are
colored with red. The speed vectors are associated to
the each dynamic entity (yellow color). The
representation result is also projected on the left
camera image (Figure 7.d). We considered that the
obstacles with a speed greater than 8km/h are
dynamic.
VisionbasedReal-timeModelingofDynamicUnstructuredEnvironmentsinDrivingScenarios
147
Figure 7: a) An urban traffic scenario. b) The alignment
result (red color) between the model delimiter extracted in
previous frame (yellow) and the data contour, extracted in
the current frame (blue). c) The virtual view of the scene.
The static obstacles are represented with green delimiters
while the dynamic obstacles are colored with red. The speed
vectors are associated to each dynamic entity. d) The
representation result, projected on the left camera image.
Figure 8 shows a comparative result between the
ICP algorithm that includes all correspondence points
(blue color) and the alignment method that uses the
Contour Boundary Rejection strategy (red color). We
used the Average Euclidean Distance (AED) as the
error metric. It can be observed that the ICP algorithm
based on Boundary Rejection strategy converge more
quickly than the ICP method without a filtering
mechanism and proves to be more accurate having a
lower alignment error. For our experiments we used a
maximum number of 10 iterations. The average
processing time was about 38 ms.
Figure 8: The computed Error Metric in the case of ICP
algorithm that does not use outlier rejection (blue color)
and ICP method that uses a Boundary Rejection (red
color).
6 CONCLUSIONS
In this paper we propose a method of real-time
representation of the dynamic environment by using
the pairwise alignment of free-form models. Instead
of registering the whole 3D point cloud, the most
visible obstacle points from the ego car are extracted
and are subjected to the alignment process. We
extend the classical ICP algorithm with a set of
preprocessing tasks. First, we associate the
delimiters at the blob level. Then, a list of associated
candidates is passed to the alignment stage. For the
registration process we use free-form polygonal
models that minimize the erroneous results caused
by occlusions, or by stereo reconstruction errors.
As future work we propose to improve the
stability of the environment perception by extending
our system with a temporal filtering of the estimated
speeds.
REFERENCES
Franke, U., Rabe, C., Badino, H., Gehrig, S., 2005. 6d-
vision: Fusion of stereo and motion for robust
ICINCO2012-9thInternationalConferenceonInformaticsinControl,AutomationandRobotics
148
environment perception. In 27th Annual Meeting of
the German Association for Pattern Recognition
DAGM ’05, pp. 216-223.
Danescu, R., Nedevschi, S., Meinecke, M.M., Graf, T.,
2007. Stereovision Based Vehicle Tracking in Urban
Traffic Environments. In Proceedings of the IEEE
Intelligent Transportation Systems Conference (ITSC
2007), Seattle, USA, 2007.
Pfeiffer, D Franke., U., 2010. Efficient Representation of
Traffic Scenes by Means of Dynamic Stixels. In
Proceedings of IEEE Intelligent Vehicles Symposium
(IEEE-IV), 2010, pp. 217-224.
Wang, C.CThorpe., CHebert., MThrun., S., Durrant-
Whyte, H., 2007. Simultaneous localization, mapping
and moving object tracking. In Proceedings of The
International Journal of Robotics Research, 26(6),
June 2007.
Prakash, S., Thomas, S., 2007. Contour tracking with
condensation/stochastic search. In Dept. of CSE. IIT
Kanpur, 2007.
Thomas de Candia., 2010. 3d tracking of dynamic objects
with a laser and a camera. In Technical report,
Autonomous System Lab, ETH Zurich, 2010.
Madhavan, R., 2002. Terrain aided localization of
autonomous vehicles. In Symposium on Automation
and Robotics in Construction, Gaithersburg,
September 01, 2002.
Yokoyama, M., Poggio, T., 2005. A Contour-Based
Moving Object Detection and Tracking. In
Proceedings of the 14th International Conference on
Computer Communications and Networks (ICCCN
'05). IEEE Computer Society, Washington, DC, USA,
271-276.
Fairfield, N., Kantor, G. A., Wettergreen, D., 2007. Real-
Time SLAM with Octree Evidence Grids for
Exploration in Underwater Tunnels. In Journal of
Field Robotics, 2007
Danescu, R., Oniga, F., Nedevschi, S., 2012. Modeling
and Tracking the Driving Environment with a Particle
Based Occupancy Grid. In proceedings of IEEE
Transactions on Intelligent Transportation Systems,
vol. 12, No. 4, December 2012, pp. 1331-1342.
Danescu, R, Oniga, F., Nedevschi, S., Meinecke, M-M.,
2009. Tracking Multiple Objects Using Particle Filters
and Digital Elevation Maps. In Proceedings of the
IEEE Intelligent Vehicles Symposium (IEEE-IV 2009),
June 2009, Xi’An, China, pp. 88-93.
Hess, J. M., 2008. Characterizing dynamic objects in 3d
laser range data. Master's thesis, Albert-Ludwigs-
Universitt Freiburg, 2008.
Fischler, M., Bolles, R., 1981. Random sample consensus:
a paradigm for model fitting with applications to
image analysis and automated cartography. In
Communications of the ACM, vol. 24, no. 6, pp. 381–
395, 1981.
Besl P., McKay, N., 1992. A method for registration of 3d
shape. In Proceedings of Trans. Pattern Analysis and
Machine Intelligence, 12(2), 1992.
Rusinkiewicz, S., Levoy, M., 2001. Efficient Variants of
the ICP Algorithm. In Proceedings of Third
International Conference on 3D Digital Imaging and
Modeling (3DIM), June 2001.
Vatavu, A., Nedevschi, S., Oniga, F., 2009. Real Time
Object Delimiters Extraction for Environment
Representation in Driving Scenarios. In Proceedings
of ICINCO-RA 2009, Milano, Italy, 2009, pp 86-93.
Woodill, J. I., Gordon, G., Buck, R., 2004. Tyzx deepsea
high speed stereo vision system. In Proceedings of the
IEEE Computer Society Workshop on Real Time 3-D
Sensors and Their Use, Conference on Computer
Vision and Pattern Recognition, 2004.
Oniga, F., Nedevschi, S., 2010. Processing Dense Stereo
Data Using Elevation Maps: Road Surface, Traffic
Isle, and Obstacle Detection. In Proceedings of IEEE
Transactions on Vehicular Technology, Vol. 59, No. 3,
March 2010, pp. 1172-1182.
Douglas, D., Peuker, T., 1973. Algorithms for the
reduction of the number of points required to represent
a digitised line or its caricature. In Proceedings of The
Canadian Cartographer, Vol 10, pp. 112-122.
Borgefors, G. 1984. Distance transformations in arbitrary
dimensions. In Proceedings of Comput. Vis. Graph.
Image Proc. 27, 321–345.
Bentley, J.L., 1975. Multidimensional binary search trees
used for associative searching. In Communications of
the ACM. 18(9):509-517, September, 1975.
Nedevschi, S., Danescu, R., Marita, T., Oniga, F., Pocol,
C., Sobol, S., Tomiuc, C. Vancea, C., Meinecke, M.
M., Graf, T., To, T. B., Obojski, M. A., 2007. A sensor
for urban driving assistance systems based on dense
stereovision. In Proceedings of Intelligent Vehicles
2007, Istanbul, June 13-15 2007, pp. 278-286.
VisionbasedReal-timeModelingofDynamicUnstructuredEnvironmentsinDrivingScenarios
149
