Object Tracking in Video Scenes

Seminar Report
Object Tracking in Video Scenes

Seminar Report
INTRODUCTION
Object tracking can be defined as the process of segmenting an object of interest from a
video scene and keeping track of its motion, orientation, occlusion etc. in order to extract useful
information.
Object tracking in video processing follows the segmentation step and is more or less
equivalent to the ‘recognition’ step in the image processing. Detection of moving objects in
video streams is the first relevant step of information extraction in many computer vision
applications, including traffic monitoring, automated remote video surveillance, and people
tracking. There are basically three approaches in object tracking. Feature-based methods aim at
extracting characteristics such as points, line segments from image sequences, tracking stage is
then ensured by a matching procedure at every time instant. Differential methods are based on
the optical flow computation, i.e. on the apparent motion in image sequences, under some
regularization assumptions. The third class uses the correlation to measure interimage
displacements. Selection of a particular approach largely depends on the domain of the problem.
Seminar Report
Applications of object tracking
Some of the important applications of object tracking are:
1. Automated video surveillance: In these applications computer vision system is
designed to monitor the movements in an area, identify the moving objects and report
any doubtful situation. The system needs to discriminate between natural entities and
humans, which require a good object tracking system.
2. Robot vision: In robot navigation, the steering system needs to identify different
obstacles in the path to avoid collision. If the obstacles themselves are other moving
objects then it calls for a real-time object tracking system.
3. Traffic monitoring: In some countries highway traffic is continuously monitored using
cameras. Any vehicle that breaks the traffic rules or is involved in other illegal act can
be tracked down easily if the surveillance system is supported by an object tracking
system.
4. Animation: Object tracking algorithm can also be extended for animation.

Seminar Report
RELATED WORK
Many researchers have tried various approaches for object tracking. Nature of the
technique used largely depends on the application domain. Some of the research work done in
the field of object tracking includes:
1. Gyaourova, C. Kamath, S. and C. Cheung has studied the block matching
technique for object tracking in traffic scenes. A motionless airborne camera is
used for video capturing. They have discussed the block matching technique for
different resolutions and complexities [5].
2. Yoav Rosenberg and Michael Werman explain an object-tracking algorithm using
moving camera. The algorithm is based on domain knowledge and motion
modeling. Displacement of each point is assigned a discreet probability
distribution matrix. Based on the model, image registration step is carried out.
The registered image is then compared with the background to track the moving
object [6].
3. A. Turolla, L. Marchesotti and C.S. Regazzoni discuss the camera model
consisting of multiple cameras. They use object features gathered from two or
more cameras situated at different locations. These features are then combined for
location estimation in video surveillance systems [7].

Seminar Report
4. One simple feature based object tracking method is explained by Yiwei Wang,
John Doherty and Robet Van Dyck. The method first segments the image into
foreground and background to find objects of interest. Then four types of features
are gathered for each object of interest. Then for each consecutive frames the
changes in features are calculated for various possible directions of movement.
The one that satisfies certain threshold conditions is selected as the position of the
object in the next frame [3].
5. Çiˇgdem Eroˇglu Erdem and Bülent San have discussed a feedback-based method
for object tracking in presence of occlusions. In this method several performance
evaluation measures for tracking are placed in a feedback loop to track nonrigid
contours in a video sequence [8].
Steps in object tracking
The process of object tracking is summarized in the block diagram below:
Basic steps in object tracking can be listed as:
 Segmentation
 Foreground / background extraction
 Camera modeling
 Feature extraction and tracking

Seminar Report
Segmentation
Segmentation is the process of identifying components of the image. Segmentation
involves operations such as boundary detection, connected component labeling, thresholding
etc. Boundary detection finds out edges in the image. Any differential operator can be used
for boundary detection [1,2]. Thresholding is the process of reducing the grey levels in the
image. Many algorithms exist for thresholding [1,2]. Refer [2] for connected component
labeling algorithms.
Seminar Report
Foreground extraction
As the name suggests this is the process of separating the foreground and background of
the image. Here it is assumed that foreground contains the objects of interest. Some of the
methods for foreground extraction are,
Use of difference images In this method we use subtraction of images in order to find objects
that are moving and those that are not. The result of the subtraction is viewed as another grey
image called difference image. Three types of difference images are defined [1].
• Absolute accumulative difference image is given by f(x,y) = f(x,y) +1 ……………….if
|g(x,y,t ) - g(x,y,t )| > T

i+1 i
• Positive accumulative difference image is given by f(x,y) = f(x,y) +1 ……………….if
g(x,y,t ) - g(x,y,t ) > T

i+1 i
• Negative accumulative difference image is given by f(x,y) = f(x,y) +1 ……………….if
g(x,y,t ) - g(x,y,t ) > T

i i+1
The following figures illustrate the three difference images.
A gap-mountain method described in [3] can then be applied to identify image blocks that
are moving and those that are not moving. The gap-mountain method works as follows- Consider
a difference image shown in the adjacent figure. A gap is a sequence of consecutive black pixels
and mountain is a sequence of consecutive white pixels. If width of a mountain in a particular

Seminar Report
row is greater than a preset threshold then we assume that a moving object is present in that row.
Similar technique is the algorithm proceeds by dividing the image into smaller sub images (or
sub matrices) until each sub matrix contains exactly one object. In the adjacent figure by
choosing proper thresholds we can detect the presence of two blocks.

Seminar Report
CONCLUSION
From the discussion, it can be seen that object tracking has many useful applications in
the robotics and computer vision fields. Several researchers have explored and implemented
different approaches for tracking. The success of a particular approach depends largely on the
problem domain. In other words, a method that is successful in robot navigation may not be
equally successful in automated surveillance. Further there exists a cost/performance trade off.
For real time applications we may need a fast high performance system on the other hand offline
applications we may use a relatively cheap (and slower in performance). It can also be seen from
the diverse nature of the techniques used that the field has a lot of room for improvement.

Object Tracking in Video Scenes

Uploaded by

Copyright:

Available Formats

Object Tracking in Video Scenes

Uploaded by

Document Information

Original Title

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

Object Tracking in Video Scenes

Uploaded by

Copyright:

Available Formats

Seminar Report

Object Tracking in Video Scenes

Applications of object tracking

Some of the important applications of object tracking are:

1. Automated video surveillance: In these applications computer vision system is

humans, which require a good object tracking system.

objects then it calls for a real-time object tracking system.

3. Traffic monitoring: In some countries highway traffic is continuously monitored using

be tracked down easily if the surveillance system is supported by an object tracking

4. Animation: Object tracking algorithm can also be extended for animation.

the field of object tracking includes:

1. Gyaourova, C. Kamath, S. and C. Cheung has studied the block matching

technique for object tracking in traffic scenes. A motionless airborne camera is

different resolutions and complexities [5].

2. Yoav Rosenberg and Michael Werman explain an object-tracking algorithm using

moving camera. The algorithm is based on domain knowledge and motion

modeling. Displacement of each point is assigned a discreet probability

3. A. Turolla, L. Marchesotti and C.S. Regazzoni discuss the camera model

location estimation in video surveillance systems [7].

changes in features are calculated for various possible directions of movement.

object in the next frame [3].

for object tracking in presence of occlusions. In this method several performance

contours in a video sequence [8].

Steps in object tracking

The process of object tracking is summarized in the block diagram below:

Basic steps in object tracking can be listed as:

 Foreground / background extraction

 Feature extraction and tracking

Segmentation is the process of identifying components of the image. Segmentation

involves operations such as boundary detection, connected component labeling, thresholding

methods for foreground extraction are,

• Absolute accumulative difference image is given by f(x,y) = f(x,y) +1 ……………….if

|g(x,y,t ) - g(x,y,t )| > T

• Positive accumulative difference image is given by f(x,y) = f(x,y) +1 ……………….if

g(x,y,t ) - g(x,y,t ) > T

• Negative accumulative difference image is given by f(x,y) = f(x,y) +1 ……………….if

g(x,y,t ) - g(x,y,t ) > T

The following figures illustrate the three difference images.

and mountain is a sequence of consecutive white pixels. If width of a mountain in a particular

choosing proper thresholds we can detect the presence of two blocks.

You might also like