|
|
Keynotes
|
Thursday,
21 October 2004
|
Time:
9:00am - 10:30am
Room: Chiang Chen Studio Theatre |
| 9:00am
Keynote Speech 1 |
Speaker:
|
Prof.
Shih-Fu Chang, PhD, FIEEE
Digital Video and Multimedia Lab, Columbia University, USA |
Topic:
|
Video
Pattern Mining |
| 9:45am
Keynote Speech 2 |
Speaker:
|
Prof.
Chin-Hui Lee, PhD, FIEEE
School of Electrical and Computer Engineering, Georgia Institute
of Technology, USA |
Topic:
|
Discovering
Knowledge in and Extracting Information from Multimedia Patterns |
| |
|
Friday,
22 October 2004
|
Time:
9:00am - 10:30am
Room: Chiang Chen Studio Theatre |
| 9:00am Keynote
Speech 3 |
Speaker:
|
Prof.
A.G. Constantinides, PhD, FIEEE
Department of Electrical and Electronic Engineering, Imperial
College, U.K. |
Topic:
|
Recent
advances in Digital Filter Design: A Design Framework from Complete
and Partial Specifications Based on Root Moments |
| 9:45am
Keynote Speech 4 |
Speaker:
|
Prof.
W.C. Siu, PhD, SrMIEEE
Centre for Multimedia Signal Processing, Department of Electronic
and Information Engineering, The Hong Kong Polytechnic University,
H.K. |
Topic:
|
Advances
in Video Coding in the Information Era |
|
|
|
|
| Topic: |
Video
Pattern Mining |
| Speaker: |
Prof.
Shih-Fu Chang, PhD, FIEEE
Digital Video and Multimedia Lab, Department of Electrical Engineering,
Columbia University, USA |
Abstract
Detection and recognition of semantic concepts and events has
been a major research challenge for multimedia indexing. Most
efforts are based on a supervised paradigm in pursuit of the
optimal features or classifiers for detecting pre-defined classes.
Such approaches sometimes suffer major problems due to the need
of extensive annotation and custom-defined target classes for
each specific domain.
To address the problems, we have been advocating a new direction
called unsupervised pattern discovery (mining).
Patterns are recurrent, predictable occurrences of one or more
entities that satisfy statistical, associative, or relational
conditions. Patterns at the feature level may signify the occurrence
of primitive concepts (e.g., recurrent passing of pedestrians).
High-level patterns may correspond to perceptual temporal events
(e.g., play-break alternations in sports) or stories from multiple
news sources forming a distinct semantic topic.
Pattern mining from heterogeneous multimedia streams offers
tremendous potential benefits while posing critical challenges.
If successful, the discovered patters can be used to summarize
large content sources, identify common threads and outliers,
and reveal distinct features that are responsible for each pattern.
However, starting with the huge space of audio-visual-textual
features and diverse structures at different time scales, finding
useful and meaningful patterns is indeed a very challenging
task.
In this talk, I will present some recent results in mining patterns
in structured video sequences (such as sports and multi-channel
broadcast news videos). Specifically, we will discuss the potential
of statistical models like Hierarchical HMM for temporal structure
mining, probabilistic latent semantic analysis for discovering
hidden concepts, and a layered dynamic mixture model for fusing
multi-modal patterns.
I will describe a recent major project, SEMANTRIX, aiming at
reconstructing semantic threads in broadcast news videos from
multiple sources. Related research issues such as news story
segmentation and near duplicate detection will also be introduced.
Evaluations using standard benchmarks (e.g., NIST TRECVID and
TDT) will be presented.
Biography
Prof. Shih-Fu Chang leads Columbia University's Digital
Video/Multimedia Lab (www.ee.columbia.edu/dvmm)
and ADVENT industry-university consortium, conducting research
in multimedia indexing, video analysis, pervasive media, and
media authentication. Notable systems developed by his group
include VisualSEEk, VideoQ, WebSEEk for image/video searching,
WebClip for networked video editing, and Sari for online image
authentication watermarking. He has applied research to various
domains, such as a medical video library, currently funded by
NSF's DLI-2 initiative, a Digital News project, and a live sports
video filtering application. His group has played an active,
leading role in developing the multimedia description schemes
in MPEG-7.
Prof. Chang is an IEEE Fellow. He served as a general co-chair
of ACM 8th Multimedia Conference 2000, a Distinguished Lecturer
of the IEEE Circuits and Systems Society, 2001-2002, and co-chair
IEEE Multimedia Conference 2004. He has received a Navy ONR
Young Investigator Award, IBM Faculty Development Award, NSF
CAREER Award, three Best Paper Awards from the IEEE, ACM, and
SPIE, and has supervised several students receiving best student
paper awards, in the areas of multimedia indexing and manipulation.
He has served on advisory or consultant positions for several
major research institutes and media technology companies.
|
|
|
|
|
| Topic: |
Discovering
Knowledge in and Extracting Information from Multimedia Patterns |
| Speaker: |
Prof.
Chin-Hui Lee, PhD, FIEEE
School of Electrical and Computer Engineering, Georgia Institute
of Technology, USA |
Abstract
In recent years, a growing population of computer users and
an exponentially increase of web pages have made the internet
the most extensive database ever established and the largest
playground ever existed. These vast varieties of internet documents
are often multilingual in nature, and consist of heterogeneous
contents, ranging from text to audio, speech, image, video,
graphics, and their combinations. Pattern intelligence is becoming
a critical research field for the representation, indexing,
retrieval, rendering, access, and presentation of media documents,
in which discovering knowledge and extracting information embedded
in multimedia patterns play key roles. In this talk, fundamental
building blocks that cut across many media dimensions are presented.
Some challenges in each technology area are then addressed.
Finally a few example applications are illustrated.
Biography
Chin-Hui Lee is currently a professor at School of
Electrical and Computer Engineering, Georgia Institute of
Technology. Dr. Lee received the B.S. degree in Electrical
Engineering from National Taiwan University, Taipei, in 1973,
the M.S. degree in Engineering and Applied Science from Yale
University, New Haven, in 1977, and the Ph.D. degree in Electrical
Engineering with a minor in Statistics from University of
Washington, Seattle, in 1981.
After graduation, Dr. Lee joined Verbex Corporation, Bedford,
MA, and was involved in research on connected word recognition.
In 1984, he became affiliated with Digital Sound Corporation,
Santa Barbara, where he engaged in research and product development
in speech coding, speech synthesis, speech recognition and
signal processing for the development of the DSC-2000 Voice
Server. Between 1986 and 2001, he was with Bell Laboratories,
Murray Hill, New Jersey, where he became a Distinguished Member
of Technical Staff and Director of the Dialogue Systems Research
Department. His research interests include multimedia communication,
multimedia signal and information processing, speech and speaker
recognition, speech and language modeling, spoken dialogue
processing, adaptive and discriminative learning, biometric
authentication, information retrieval, and bioinformatics.
His research scope is reflected in "Automatic Speech
and Speaker Recognition: Advanced Topics", published
by the Kluwer Academic Publishers in 1996. From August 2001
to August 2002 he was a visiting professor at School of Computing,
The National Univ. of Singapore. In Sept. 2002, he joined
the Faculty of School of Electrical and Computer Engineering
at Georgia Institute of Technology.
Prof. Lee has participated actively in professional societies.
He is a member of the IEEE Signal Processing Society, Communication
Society, and the European Speech Communication Association.
He is also a lifetime member of the Computational Linguistics
Society in Taiwan. In 1991-1995, he was an associate editor
for the IEEE Transactions on Signal Processing and Transactions
on Speech and Audio Processing. During the same period, he
served as a member of the ARPA Spoken Language Coordination
Committee. In 1995-1998 he was a member of the Speech Processing
Technical Committee of the IEEE Signal Processing Society
(SPS), and later became the chairman of the Speech TC from
1997 to 1998. In 1996, he helped promote the SPS Multimedia
Signal Processing (MMSP) Technical Committee in which he is
a founding member.
Dr. Lee is a Fellow of the IEEE, and has published more than
250 papers and 25 patents on the subject of automatic speech
and speaker recognition. He received the SPS Senior Award
in 1994 and the SPS Best Paper Award in 1997 and 1999, respectively.
In 1997, he was also awarded the prestigious Bell Labs President's
Gold Award for his contributions to the Lucent Speech Processing
Solutions product. In 2000, he was named one of the six Distinguished
Lecturers by the IEEE Signal Processing Society.
|
|
|
|
|
| Topic: |
Recent
advances in Digital Filter Design: A Design Framework from Complete
and Partial Specifications Based on Root Moments |
| Speaker: |
Prof.
A.G. Constantinides, PhD, FIEEE
Department of Electrical and Electronic Engineering, Imperial
College, U.K. |
Abstract
The talk is concerned with the following problem. Given a table
of numbers that represent requirements either in amplitude A(q),
phase f(q)
or group delay response t(q),
and which do not necessarily cover the entire frequency range,
but are limited over a set of frequencies {q1,
q2},
to design a transfer function H(z) which satisfies one set of
specifications. H(z) is further constrained to be either a finite
impulse response (FIR) or autoregressive (AR) transfer function.
The problem is further extended to cover the combined satisfaction
of any pairing of A(q)
and either f(q)or
t(q),
in which any set of specifications may be partially specified.
The presentation proposes a framework to deal with these problems
and suggests indicative solutions. The solutions proposed are
based on the use of root moments. The subordinate problem concerning
the estimation of the order of the FIR or AR filter from partial
specifications on A(q)
or f(q)
or t(q)is
also examined. An additional intention of the presentation is
to stimulate further research into novel digital filter design
methods. For this purpose there are interspersed suggestions
for further examination and study.
Biography
A. G. Constantinides is the Professor of Signal Processing
and the Head of the Signal Processing and Digital Systems
Section of the Department of Electrical and Electronic Engineering
at Imperial College, London, U.K. He has been actively involved
with research in various aspects of digital filter design,
digital signal processing, and communications for more than
30 years. Professor Constantinides' research spans a wide
range of Digital Signal Processing, from the theoretical,
as well as the practical, points of view. His recent work
has been directed toward the demanding signal processing problems
arising from the various areas of telecommunication. This
work is supported by research grants and contracts from various
government and industrial organisations. He has published
several books and papers in learned journals in the area of
Digital Signal Processing and its applications. Professor
Constantinides has served as the First President of the European
Association for Signal Processing (EURASIP). He has been on,
and is currently serving as, a member of many technical program
committees of the IEEE, the IEE and other international conferences.
He has organized the first international series of meetings
on Digital Signal Processing, initially in London, in 1967,
and in Florence since 1972. In 1985, he was awarded the Honour
of Chevalier, Palmes Academiques, by the French government,
and in 1996, the promotion to Officier, Palmes Academiques.
He holds honorary doctorates from European and Far Eastern
Universities, several Visiting Professorships, Distinguished
Lectureships, Fellowships and other honors around the world.
He served as a Member of the Board of Governors of the IEEE
Signal Processing Society, a member of several Technical Committees
of the IEEE and the Institution of Electrical Engineers (IEE),
U.K., and is on the Editorial Boards of many professional
journals. He is a Fellow of IEE and an honorary member of
Eta Kappa Nu.
|
|
|
|
|
| Topic: |
Advances
in Video Coding in the Information Era |
| Speaker: |
Prof.
W.C. Siu, PhD, SrMIEEE
Centre for Multimedia Signal Processing, Department of Electronic
and Information Engineering, The Hong Kong Polytechnic University,
H.K. |
Abstract
Video coding becomes an indispensable part of modern multimedia
and information systems. The development of video codec started
from simple hybrid video coding, and gradually diverges to the
object-oriented video coding and then the optimized high quality
and low bit rate video coding. The approaches become a trade-off
between the quality and resources available. In this presentation,
we begin to trace through the primary reasons for the advancement
of video coding and give examples of the related technologies.
Subsequently, some highlights of the recent research results
on partial distortion search, video transcoding, wavelet videos,
etc. and samples of our recent studies in the Centre for Multimedia
Signal Processing will be presented. A number of illustrations
and demonstrations will be given to show the basic principles
and results of our realization. The presentation will end with
some ideas on new trends and directions in the area.
Biography
Wan-Chi Siu received the Associateship from the Hong
Kong Polytechnic University, the MPhil degree from the Chinese
University of Hong Kong and the PhD degree from the Imperial
College of Science, Technology & Medicine, London, in
1975, 1977 and 1984, respectively. He joined the Hong Kong
Polytechnic University as a lecturer in 1980 and has been
Chair Professor of the Department of Electronic & Information
Engineering since 1992. He has published over 250 research
papers, over 120 of which appeared in international journals,
such as IEEE Transactions and Signal Processing. His research
interests include digital signal processing, fast computational
algorithms, transforms, wavelets, image and video coding,
and computational aspects of pattern recognition and neural
networks. Professor Siu was Guest Editor and Associate Editor
of the IEEE Transactions on Circuits and Systems, Pt.II between
1995-97. He is also an editor of the recent book, 'Multimedia
Information Retrieval and Management', Springer Berlin Heidelberg
2003, and a member of the editorial board of the Journal of
VLSI Signal Processing Systems for Signal, Image, Video Technology,
the EURASIP Journal on Applied Signal Processing, Pattern
Recognition and a few other journals. Prof. Siu received many
awards, such as the Distinguished Presenter Award of the University
Management Workshop of the PolyU (1997), IEEE Third Millennium
Medal for Outstanding Achievements and Contributions (2000),
IEEE USA, and recently the Best Teacher Award (2003), EIE,
the Outstanding Award in Research & Scholarly Activities,
Faculty of Engineering, PolyU (2003), Plaque for Exceptional
Leadership and Significant Contributions, IEEE SP Society
Conference Board (2003), IEEE USA, and Honorable Mention Winner
of the 30th Annual Pattern Recognition Society Award (2004),
USA. He has been a keynote speaker of many international conferences,
including recently the 2002 Third IEEE Pacific Rim Conference
on Multimedia (PCM2002), Taiwan, and the 2003 IEEE International
Conference on Neural Networks and Signal Processing, Nanjing.
Professor Siu has held the position of general chair or technical
program chair of many international events, including IEEE
society sponsored flagship conferences, such as the Technical
Program Chair of IEEE International Symposium on Circuits
and Systems (ISCAS'97) and the General Chair of 2003 IEEE
International Conference on Acoustics, Speech and Signal Processing
(ICASSP'2003). Between 1991 and 1995, Prof. Siu was a member
of the Physical Sciences and Engineering Panel of the Research
Grants Council (RGC), Hong Kong Government, and in 1994 he
chaired the first Engineering and Information Technology Panel
to assess the research quality of 19 Cost Centers (departments)
from all universities in Hong Kong.
|
|
|
|