Download (44.5k) - IEEE Computer Society

lynxherringΤεχνίτη Νοημοσύνη και Ρομποτική

18 Οκτ 2013 (πριν από 3 χρόνια και 9 μήνες)

94 εμφανίσεις

Speaker Information


Name of Nominee: Sharath Pankanti


Organization: IBM T J Watson Research Center


Mailing Address: 1101 Kitchawan Road, Rte 134, Yorktown Heights NY 10598


Phone: 914 945 1500


Fax: (914) 945 4252


Email: sharat@us.ibm.com


Include with

this nomination the following information. Electronic submissions are


preferred. Nominees are required to submit a brief biography, and 2
-
3 abstracts of talks


to be given which fit on one page in the DVP web site. (See current speaker page for


example)


Biography


Sharath Pankanti received PhD in Computer Science from Michigan State University.

He is manager of Exploratory Computer Vision Group at T J Watson Research Center,

IBM. He leads a number of safety, productivity, and security focused projects i
nvolving

biometrics
-
, multi
-
sensor surveillance, driver assistance technologies that entail object/

event modeling, detection and recognition from information provided by static and

moving sensors and cameras. Many of these works are integrated into system
s that

have been rigorously evaluated in real world applications including the prestigious

Frost and Sullivan award winning IBM surveillance offering that has been featured in

mainstream media including ABC and Fox. His research interests include performan
ce

metrics metrics/evaluation, computer vision system designs for effective privacy, safety,

security, productivity, and convenience. He has published over 125 publications in peer
-

reviewed conference/workshop proceedings and journals and has contributed
to over 80

inventions spanning biometrics, object detection, and recognition which has resulted in

significant impact (see google scholar, ms academic, DBLP profiles). He has co
-
edited

the first comprehensive book on biometrics, "Biometrics: Personal Ident
ification" Kluwer,

1999 and co
-
authored, "A Guide to Biometrics", Springer 2004 which is being used in

many undergraduate and graduate biometrics curricula. Sharath is a fellow of IEEE and

served the computer vision and pattern recognition community in man
y over last two

decades.


Sample Abstracts


Title 1

Actionable Video Content Summarization: Lessons from Practical Case Studies.


For the first time in the history of universe, the video data generated by humans has

exceeds all other forms of the data. Whi
le no one denies the utility of the information

hidden within this deluge of data, it is clear that it is humanly impossible to browse,

navigate, and search a deluge of data from cameras and other sensors in a variety of

applications including retail surve
illance, railroad inspection, unmanned aerial vehicle

(UAV) video surveillance. To effectively find actionable information from these massive

streams of data is an important question with both scientific and business value. The

practical systems that we bu
ilt, although in pursuit of different business objectives, share

a common goal, which is to intelligently and efficiently analyze and extract the most

important actionable information from an overwhelming amount of data, while being able

to effectively ign
ore a large portions of uneventful and/or noisy data. I will summarize a

variety of computer vision, machine learning and system optimization techniques that we

used to successfully address different technical and business challenges, and to deliver

differ
entiating performance to meet our customers' expectations.


Joint work with Lisa Brown, Jon Connell, Ankur Datta, Quanfu Fan, Rogerio Feris,

Norman Haas, Jun Li, , Ying Li, : Sachiko Miyazawa, Juan Moreno, Hoang Trinh.


Title 2 Video Analytics and Public S
afety


The remarkable growth of sensor data acquisition has lead to a situation wherein

we have a shortage of personnel to monitor all of the data that is being generated.

Such a scenario is typical in a video surveillance situation, where a massive number

of cameras are being deployed to monitor large geographical areas, such as cities. A

typical municipal command and control center will have camera monitors covering an

entire wall and a bevy of humans monitoring all the incoming video feeds for suspicious

activities. Such human monitoring not only suffers from loss of attentiveness, since one

cannot simultaneously focus on all the activities in all the cameras at once, but also from

human fatigue and boredom while looking at these camera feeds for extended

periods of

time.

Typical surveillance systems offer not only real
-
time alerting capabilities, but also enable

the user to search for events of interest after the fact for upto hundreds of cameras.

Many commercial systems for intelligent urban surveillance

exist in the These systems

have accomplished practically a lot in last two decades, especially, when we consider

the capabilities of manual video surveillance systems or the cost of “cop on the beat”

approach. This talk will overview accomplishments of mo
dern surveillance systems and

outstanding technical challenges.


Joint work with Lisa Brown, Jon Connell, Ankur Datta, Quanfu Fan, Rogerio Feris,


Chiao
-
Fe Shu, Arun Hampapur.


Title 3: Retail Video Analytics


The application of computer vision techniques
in retail dates back more than a decade. .

Today retail video analytics has gone beyond the traditional domain of security and loss

prevention by providing retailers insightful business intelligence. Such information allows

for enhanced customer experience
, optimized store performance, reduced operational


costs, and ultimately higher profitability. Due to advances in computer vision, machine

learning, and data analysis, retail video analytics can provide retailers with much more

insightful business intelli
gence. Thus it promises much higher business value, far

beyond the traditional domain of security, authentication, and loss prevention. Examples

include analysis of store traffic, queue data, shoppers’ behaviors, and purchase decision

making, among others.

The retail environment, with its own unique business and

technical challenges, is also considered a practical test bed for novel computer vision

approaches. For these reasons, retail video analytics and its various applications have

become of great intere
st to both retailers as well as the computer vision community. This

talk provides an overview of various camera
-
based applications in retail as well as the

state
-
of
-
the
-
art computer vision techniques behind them. It also presents some of the

promising tech
nical directions for exploration in retail video analytics.


Joint work with J. Connell, Q. Fan, Prasad Gabbur, Norman Haas, Hoang Trinh.


Title 4: Grand Challenges in Biometrics


Reliable person recognition is an important problem in diverse businesses. B
iometrics,

recognition based on distinctive personal traits, has the potential to become an

irreplaceable part of many identification systems. While successful in some niche

markets, the biometrics technology has not yet delivered its promise of foolproof

automatic human recognition. With the availability of inexpensive biometric sensors

and computing power, it is becoming increasingly clear that broader usage of biometric

technologies is being stymied by our lack of understanding of four fundamental

proble
ms: (i) How to accurately and efficiently represent and recognize biometric

patterns? (ii) How to guarantee that the sensed measurements are not fraudulent? (iii)

How to make sure that the application is indeed exclusively using pattern recognition for

the

expressed purpose (function creep)? (iv) How to acquire repeatable and distinctive

patterns from a broad population? Solving these core problems will be required to move

biometrics into mainstream applications and may also stimulate adoption of other patt
ern

recognition applications for providing effective automation of sensitive tasks without

jeopardizing individual freedoms. For these reasons, we view biometrics as a grand

challenge
-

"a fundamental problem in science and engineering with broad economic

and scientific impact”.


Joint work with Anil K. Jain, Salil Prabhakar, Lin Hong, Arun Ross, James L. Wayman


Person Submitting Nomination


Name: Sharath Pankanti (self)


Organization: IBM T J Watson Research Center


Mailing Address: 1101 Kitchawan Rd. Yor
ktown Heights NY 10598.


Phone: 914 945 1500


Fax: (914) 945 4252


Email: sharat@us.ibm.com


Selection Criteria Summary


*Prior Invited Presentations


1. S. Pankanti et al. Actionable Video Content Summarization: Lessons from

Practical Case Studies, The Se
cond International conference on Intelligent

Interactive Technologies and Multimedia”, March 08
-
11, 2013, Allahabad, India.

Expected Audience size: ~100

2. Building Practical Vision Systems: Case Studies in Surveillance and Inspection,

The First Internatio
nal Conference on Multimedia Processing, Communication

and Computing Applications, Bangalore 13
-
15 December, 2012. Expected

Audience size: ~100

3. Vision
-
based Integrated Systems for Intelligent Data Analysis in Surveillance

and Inspection, International C
onference on Advanced Computer Science and

Information Systems 2012, Universitas Indonesia, December 1
-
2, 2012, 2012.

Audience size: ~50

4. S. Pankanti, Q. Fan , Hoang Trinh, J, Connell, N. Haas, Video Analytics: Public

Sector Surveillance. International C
onference on Communication, Computation,

Management and Nanotechnology, Bhalki, India, Sep 23, 2011. Audience size:

~100

5. S. Pankanti, R. Bobbitt, J. Connell, Q. Fan, R. Feris, P.Gabbur, N. Haas, A.

Hampapur, C. Otto, , U. Park, D. Tran, H. Trinh, Y. Zha
i, Visual Analytics for

Shrinking Checkout Shrink, ECCV, Crete, Greece, Sept 5, 2010. . Audience

size: ~100

6. S. Pankanti, R. Bobbitt, J. Connell, Q. Fan, R. Feris, P.Gabbur, N. Haas, A.

Hampapur, C. Otto, , U. Park, D. Tran, H. Trinh, Y. Zhai, Challenges

of using

Video Analytics for Detection of Shoplifting Behavior, Video Surveillance

Technologies for Retail Security Symposium, NIST, Gaithersburg, MD, June 29
-

30, 2010. . Audience size: ~100

7. S. Pankanti, Q. Fan, Y. Zhai, R. Bobbitt, A. Yanagawa, S. Mi
yazawa,

R. Kjeldsen, A. Hampapur, Multi
-
media Compliance: A practical paradigm

for managing business integrity Emerging Multimedia Circuits and Systems

Technologies, ICME, July 1, 2009, New York, New York.. Audience size: ~100


*Articles in Trade Journals


1. R. Feris, B. Siddiquie, J. Petterson, Y. Zhai, A. Data, L. Brown, S. Pankanti.

Large
-
Scale Vehicle Detection, Indexing, and Search in Urban Surveillance

Videos, Object and Event Classification in Large
-
Scale Video Collections,

IEEE Transactions on Mult
imedia
-

Special Issue, pp. 99, 1
--
1, IEEE, 2012.


2. S. Pankanti, L. Brown, J. Connell, A. Datta, Q. Fan, R. Feris, N. Haas, Y.

Li, N. Ratha, H. Trinh, Practical computer vision: Example techniques and

challenges, IBM Journal of Research and Development C
entennial Issue:

Sept.
-
Oct. 2011 Volume : 55 , Issue:5 On page(s): 3:1
-

3:12.

3. A. Hampapur, H. Cao, A. J. Davenport, W. S. Dong, D. Fenhagen, R.

S. Feris, G. Goldszmidt, Z.B. Jiang, J. Kalagnanam, T. Kumar, H. Li,

S. Mahatma, S. Pankanti, D. Pelleg, W.
Sun, M. Taylor, C. H. Tian, S.

Wesserkrug, L. Xie, M. Lodhi, C. Kiely, K. Butturff, L. Desjardins, Analytics
-

driven asset management, IBM Journal of Research and Development.

2011, 13
-
13.

4. A. K. Jain and S. Pankanti, Beyond Fingerprinting, Scientific Am
erican, 2008.

5. K. Nandakumar, A. K. Jain and S. Pankanti, Fingerprint
-
based Fuzzy

Vault: Implementation and Performance, IEEE Transactions on Information

Forensics and Security, 2008. (winner of IEEE SPS 2011 young author Best

Paper Award).

6. A. K. Jain

and S. Pankanti, A Touch of Money, IEEE Spectrum, pp. 22
-
27,

July 2006.

7. A. K. Jain, A. Ross and S. Pankanti, Biometrics: A Tool for Information

Security, IEEE Transactions on Information Forensics and Security Vol. 1,

No. 2, pp. 125
-
143, June 2006.

8.
A. W. Senior, A. Hampapur, Y
-
L. Tian, L. Brown, S. Pankanti, R. M. Bolle:

Appearance models for occlusion handling. Image Vision Comput. 24 (11):

1233
-
1243 (2006).

9. A. W. Senior, S. Pankanti, A. Hampapur, L. Brown, Y
-
L. Tian, A. Ekin, J. H.

Connell, C
-
F.

Shu, M.Lu: Enabling Video Privacy through Computer Vision.

IEEE Security & Privacy 3 (3): 50
-
57 (2005).

10. A. Hampapur, L. M. Brown, J. Connell, M. Lu, H. Merkl, S. Pankanti, A.

W. Senior, C
-
F. Shu, and Y
-
L. Tian, Multiscale Tracking for Smart Video

Surv
eillance, IEEE Trans. Signal Processing, Volume 22, no. 2, 2005, pp.

38

51.

11. U. Uludag, S. Pankanti, S. Prabhakar, A. K. Jain, Biometric Cryptosystems:

issues and challenges Proceedings of the IEEE Volume: 92 , Issue: 6 , June

2004 Pages: 948
-

960.

12.

R. Bolle, N. Ratha, S. Pankanti, Error analysis of pattern recognition systems

the subsets bootstrap, Computer Vision and Image Understanding (CVIU)

Journal. Volume 93, Issue 1, January 2004, Pages 1
-
33.

13. S. Prabhakar, A. K. Jain, and S. Pankanti, Lear
ning Fingerprint Minutiae

Location and Type, Pattern Recognition, Volume 36, No. 8, pp. 1847
-
1857,

2003.

14. S. Prabhakar, S. Pankanti, A. K. Jain, Biometric Recognition: Security and

Privacy Concerns, IEEE Security and Privacy, Volume 1. No. 2,pp. 33
-
42,

March
-
April 2003.

15. S. Pankanti, A. Senior, L. Brown, A. Hampapur, Y. Tian, R. Bolle,

Peoplevision: Privacy Protection in Visual Surveillance, IEEE Pervasive

Computing Magazine, p. 96, February 2003.

16. S. Pankanti, S. Prabhakar, A. K. Jain, On Individu
ality of Fingerprints,

Fingerprint Whorld, pp. 150
-
159, July 2002.

17. S. Pankanti, S. Prabhakar, A. K. Jain, On the individuality of fingerprints,

IEEE Transactions on Pattern Analysis and Machine Intelligence, July 2002,

pp 152
-
159.


18. A. K. Jain, S. P
rabhakar, and S. Pankanti, On The Similarity of Identical Twin

Fingerprints, Pattern Recognition, Vol. 35, No. 11, pp. 2653
-
2663, 2002.

19. A. K. Jain and S. Pankanti, Biometrics Systems: Anatomy of Performance,

IEICE Trans. Fundamentals, Vol. E84
-
D, No. 7
, pp. 788
-
799, 2001.

20. K. Jain, S. Prabhakar, S. Pankanti, Matching and Classification: A Case

Study in Fingerprint Domain, Proc. INSA
-
A (Indian National Science

Academy), December 2000.


* Books Published (edited and authored)


1. P. J. Flynn; S. Pankan
ti (Eds), Proceedings of SPIE Conference on Biometric

Technology for Human Identification III Proceedings of SPIE Volume: 6202

April 2006

2. S. Z. Li, Z. Sun, T. Tan, S. Pankanti, G. Chollet, D. Zhang (eds): Advances

in Biometric Person Authentication, Int
ernational Workshop on Biometric

Recognition Systems, IWBRS2005, Beijing, China, October 22
-
23, 2005,

Proceedings LNCS Springer 2005. vol. 3781, Springer, Berlin, 2005.

3. R. Bolle, J. Connell, S. Pankanti, A. Senior, N. Ratha, Guide to Biometrics,

Spriger
, 2003. (translated into Polish and Russian)

4. Jain, R. Bolle, and S. Pankanti (Eds.), Biometrics: Personal Identification in

Networked Society, Kluwer Academic, 1999.


* Topics of Interest to Computer Society Members


Computer Vision Applications, Video
Analytics, Video Surveillance, Multimodal sensor


integration, machine learning, Driver Assistance systems, Rail Inspection, Eldercare,


Retail video analytics. Biometrics, Privacy, Person Identification, Aerial Surveillance.


IMPORTANT: Nominations for ne
w speakers are welcome at any time and can be made


by anyone. All nominations received by 1 November will be considered for the program


year beginning in January. Nominee materials are reviewed and new speakers are


approved in December.


Send completed
nominations to John Daniel at jw.daniel@computer.org


The Distinguished Visitors Program (DVP)


c/o IEEE Computer Society


2001 L Street, NW, Suite 700


Washington, DC 20036


Phone: +1 202 371 0101


+1 202 371 0101 FREE