IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 19, NO.
10, OCTOBER 2010
Special Section on Distributed Camera Networks:
Sensing, Processing, Communication, and
Implementation
R ECENT advances in capture, processing, and commu-
nications hardware have resulted in the proliferation of
small, versatile, and inexpensive network cameras that can be
easily installed and remotely managed. Millions of network
cameras have already been deployed by federal and local gov-
ernments, educational institutions, and private businesses. In
a traditional setup, all cameras stream to a central location for
processing; however, this model does not scale well with the
number of cameras. Moreover, even if centralized processing
were sustainable, separate processing of each camera’s stream
may be insufficient for the detection and recognition of com-
plex, wide-area events. Thus, there is a need for distributed
algorithms, such as in-camera (edge) processing, which lead
to reduced bandwidth requirements, as well as collaborative
processing, which permits joint decision making and is of
particular interest in the case of complex events, occlusions,
extreme lighting conditions, etc.
The call for papers for this Special Section elicited an en-
thusiastic response. After a rigorous review process, eight pa-
pers were accepted for publication. The eight papers comprising
this issue cover diverse topics on calibration, power-constrained
adaptive sensing, tracking, activity recognition, and fusion using
manifolds.
Because a typical camera network is not calibrated (camera
location and scene layout are both unknown), many well-un-
derstood algorithms from the literature do not apply. Lobaton,
Vasudevan, Bajcsy, and Sastry propose a camera-network repre-
sentation that accurately captures topological information about
the visual coverage of the network. This representation pro-
vides coordinate-free calibration of the network and does not
require localization of the cameras or objects in the environ-
ment. The authors also propose a distributed, robust algorithm
for the construction of this representation using only binary de-
tection information.
Many applications involving camera networks require accu-
rate calibration of sensors. The paper by Kassebaum, Bulusu,
and Feng describes a novel calibration algorithm using a known
3-D target full of features visible to all the cameras. This algo-
rithm is suitable for power-constrained environments.
Power-consumption constraints for video transmission in
wireless systems can be prohibitive when using video sensors
with limited power resources. For instance, various transmis-
sion schemes can be used to significantly alter the lifetime of
battery-operated video sensors. Marijan, Demirkol, Maričić,
Sharma, and Ignjatović propose an optimal power-alloca-
tion strategy for wireless video sensors using a pixel-level
Digital Object Identifier 10.1109/TIP.2010.2063650
1057-7149/$26.00 © 2010 IEEE
2513
sigma-delta image sensor. Optimization is achieved by min-
imizing the power-rate-distortion characteristics of the video
sensor, thus, relying on adaptive sensing to reach an optimal
balance between sensing, compression, and transmission. This
approach has been demonstrated to improve the video quality
and prolong the lifetime of the sensor.
Yoder, Medeiros, Park, and Kak present a distributed
multi-camera face tracking system that does not require a
central server to coordinate the entire tracking effort. They use
a clustering protocol to dynamically form groups of cameras
for in-network tracking of individual faces. The experimental
results show that the cluster-based distributed face tracker is
capable of accurately tracking multiple faces in real-time.
On the other hand, Song, Kamal, Soto, Ding, and
Roy-Chowdhury investigate distributed scene analysis using
concepts of consensus from multiagent systems. While each
camera estimates certain parameters independently, these pa-
rameters are subsequently shared with neighboring cameras to
arrive at a final estimate. The authors focus on two basic surveil-
lance tasks in this context: tracking and activity recognition.
One way to cope with the voluminous data collected by a
camera network is to develop low-dimensional data models.
Manifold models provide a particularly powerful approach for
capturing the structure of data governed by a low-dimensional
set of parameters. An important consideration is to be able to ac-
count for dependencies among multiple sensors. The paper by
Davenport, Hegde, Duarte, and Baraniuk proposes a new joint
manifold framework that exploits such dependencies. Using the
recently developed results concerning random projections of
manifolds, the authors present a network-scalable dimension-
ality reduction scheme that efficiently fuses the data from all
sensors.
Appearance-based matching in a camera network is not ro-
bust to illumination and view changes. The paper by Ermis,
Clarot, Jodoin, and Saligrama presents an approach that uses ac-
tivity features for matching across cameras. The basic premise
of this paper is that similar activity features will be generated
by a given activity irrespective of where and when it occurs.
The challenge is in identifying and extracting such features. The
paper suggests using time-series features generated by the ac-
tivity under consideration.
The paper by Wang, Velipasalar, and Casares concentrates
on the practical aspect of cooperative tracking and composite
event detection across multiple cameras. More specifically,
the authors present lightweight, robust and bandwidth-efficient
(peer-to-peer) algorithms that have been implemented on a
wireless, embedded platform employing the CITRIC mote.
This type of implementation promises camera network scala-
bility under power and bandwidth constraints, issues critical to
2514 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 19, NO. 10, OCTOBER 2010

autonomous, efficient and effective video surveillance in the as well as publications coordinator, Ms. Kathy Jackson, for their
future. leadership and support of this Special Section.
The papers included in this Special Section span across
theoretical and practical considerations of various aspects RAMA CHELLAPPA, Guest Editor
of distributed camera networks, including adaptive sensing, University of Maryland
distributed processing, efficient communications, and versatile College Park, MD 20742
implementations. These papers serve as a seed for many new
and exciting developments that will emerge over the next WENDI HEINZELMAN, Guest Editor
decade in the rapidly developing area of distributed camera University of Rochester
networks. Rochester, NY 14627
We thank the authors of all submitted papers for their con-
tributions; we would also like to express our appreciation to all JANUSZ KONRAD, Guest Editor
of the reviewers for their comments and suggestions. We would Boston University
finally like to extend our gratitude to the current and former Ed- Boston, MA 02215
itors-in-Chief, Prof. Charles Bouman and Prof. Thrasos Pappas,
DAN SCHONFELD, Guest Editor
University of Illinois at Chicago
Chicago, IL 60607

MARILYN WOLF, Guest Editor

Georgia Institute of Technology
Atlanta, GA 30332

Rama Chellappa (S’78–M’79–SM’83–F’92) received the B.E. (Hons.) degree from the University
of Madras, India, in 1975, and the M.E. (Distinction) degree from the Indian Institute of Science,
Bangalore, in 1977. He received the M.S.E.E. and Ph.D. Degrees in Electrical Engineering from
Purdue University, West Lafayette, IN, in 1978 and 1981, respectively.
Since 1991, he has been a Professor of electrical engineering and an affiliate Professor of com-
puter science at the University of Maryland, College Park. He is also affiliated with the Center
for Automation Research (Director) and the Institute for Advanced Computer Studies (Permanent
Member). In 2005, he was named a Minta Martin Professor of Engineering. Prior to joining the
University of Maryland, he was an Assistant (1981-1986) and Associate Professor (1986-1991)
and Director of the Signal and Image Processing Institute (1988-1990) at University of Southern
California, Los Angeles. Over the last 29 years, he has published numerous book chapters, peer-re-
viewed journal, and conference papers. He has coauthored and edited many books in visual surveil-
lance, biometrics, MRFs and image processing. His current research interests are face and gait
analysis, 3-D modeling from video, image and video-based recognition, and exploitation and hyper spectral processing.
Prof. Chellappa has served as an Associate Editor of four IEEE TRANSACTIONS, as a Co-Editor-in-Chief of Graphical
Models and Image Processing, and as the Editor-in-Chief of the IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE
INTELLIGENCE. He served as a member of the IEEE Signal Processing Society Board of Governors and as its Vice President of
Awards and Membership. He is serving a two-year term as the President of IEEE Biometrics Council. He has received several
awards, including an NSF Presidential Young Investigator Award, four IBM Faculty Development Awards, an Excellence in
Teaching Award from the School of Engineering at USC, two paper awards from the International Association of Pattern
Recognition, the Technical Achievement and Meritorious Service Awards from the IEEE Signal Processing Society, the
IEEE Computer Society. At the University of Maryland, he was elected a Distinguished Faculty Research Fellow and a
Distinguished Scholar-Teacher; he received the Outstanding Faculty Research Award from the College of Engineering, received
an Outstanding Mentor Award from the GEMSTONE program, and an Outstanding Innovator Award from the Office of
Technology Commercialization. He is a Fellow the International Association for Pattern Recognition and the Optical Society of
America. He has served as a General the Technical Program Chair for several IEEE international and national conferences and
workshops. He is a Golden Core Member of the IEEE Computer Society and served a two-year term as a Distinguished Lecturer
of the IEEE Signal Processing Society.
IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 19, NO. 10, OCTOBER 2010 2515

Wendi Heinzelman received the B.S. degree in electrical engineering from Cornell University,
Ithaca, NY, in 1995 and the M.S. and Ph.D. degrees in electrical engineering and computer science
from the Massachusetts Institute of Technology, Cambridge, in 1997 and 2000, respectively.
She is currently an Associate Professor in the Department of Electrical and Computer Engi-
neering at the University of Rochester, NY, and holds a secondary appointment as an Associate
Professor in the Department of Computer Science. She also currently serves as Dean of Graduate
Studies for Arts, Sciences, and Engineering. Her current research interests lie in the areas of wire-
less communications and networking, mobile computing, and multimedia communication.
Dr. Heinzelman received the NSF CAREER award in 2005 for her research on cross-layer ar-
chitectures for wireless sensor networks, and she received the ONR Young Investigator Award in
2005 for her work on balancing resource utilization in wireless sensor networks. She is an As-
sociate Editor for the IEEE TRANSACTIONS ON MOBILE COMPUTING, an Associate Editor for the
ACM Transactions on Sensor Networks and an Associate Editor for Elsevier Ad Hoc Networks
Journal. She is a senior member of the ACM and she is co-founder of the N Women (Networking Networking Women) group.

Janusz Konrad (M’93–SM’98–F’08) received the M.Eng. degree from the Technical University
of Szczecin, Poland in 1980, and the Ph.D. degree from McGill University, Montreal, Canada in
1989.
From 1989 to 2000, he was with INRS-Telecommunications, Montŕeal. Since 2000, he has been
with Boston University. His research interests include image and video processing, visual sensor
networks, and 3-D visualization.
Dr. Konrad is an Associate Technical Editor for the IEEE Communications Magazine and As-
sociate Editor for the EURASIP International Journal on Image and Video Processing. He was
an Associate Editor for the IEEE TRANSACTIONS ON IMAGE PROCESSING and the IEEE SIGNAL
PROCESSING LETTERS, member of the IMDSP Technical Committee of the IEEE Signal Processing
Society, as well as the Technical Program Co-Chair of ICIP-2000, Tutorials Co-Chair of ICASSP-
2004, and Technical Program Co-Chair for AVSS-2010. He is a co-recipient of the 2001 Signal Pro-
cessing Magazine award for a paper co-authored with Dr. Ch. Stiller and the 2004-2005 EURASIP
Image Communications Best Paper Award for a paper co-authored with Dr. N. Bozinovic.

Dan Schonfeld (M’90–SM’05–F’10) received the B.S. degree in electrical engineering and com-
puter science from the University of California at Berkeley, and the M.S. and Ph.D. degrees in
electrical and computer engineering from The Johns Hopkins University, Baltimore, MD, in 1986,
1988, and 1990, respectively.
In 1990, he joined the University of Illinois at Chicago, where he is currently a Professor in the
Departments of Electrical and Computer Engineering, Computer Science, and Bioengineering. He
has authored over 170 technical papers in various journals and conferences. His current research
interests are in multidimensional signal processing; image and video analysis; computer vision;
and genomic signal processing.
Dr. Schonfeld was coauthor of papers that won the Best Student Paper Awards in Visual Commu-
nication and Image Processing 2006 and the IEEE International Conference on Image Processing
2006 and 2007. He elevated to the rank of IEEE Fellow “for contributions to image and video
analysis.” He is currently Deputy Editor-in-Chief of the IEEE TRANSACTIONS ON CIRCUITS AND
SYSTEMS FOR VIDEO TECHNOLOGY and Special Sections Area Editor for the IEEE Signal Processing Magazine.

Marilyn Wolf (S’78–M’80–SM’91–F’98) received the B.S., M.S., and Ph.D. degrees in electrical
engineering from Stanford University, Stanford, CA, in 1980, 1981, and 1984, respectively.
She is the Farmer Distinguished Chair and Georgia Research Alliance Eminent Scholar at the
Georgia Institute of Technology, Atlanta. She was with AT&T Bell Laboratories from 1984 to 1989.
She was on the faculty of Princeton University from 1989 to 2007. Her research interests included
embedded computing, embedded video and computer vision, and VLSI systems.
Dr. Wolf received the ASEE Terman Award and IEEE Circuits and Systems Society Education
Award. She is a Fellow of the ACM and an IEEE Computer Society Golden Core member.