Abstract:

     As the amount of video data increases, organizing and 
retrieving video data based on their semantics is becoming 
increasingly important. Traditionally, supervised learning 
is used to build models for detecting semantic concepts. 
However, in order to obtain a substantial amount of 
training data, extensive labeling work is needed with the 
supervised learning schemes. In this talk, we will describe 
our proposed Autonomous Learning framework, in which 
imperfect labeling automatically extracted from cross-
modality information is used for training. This completely 
avoids the manual labeling process. In the framework, 
imperfect labels without user involvement are first 
obtained from cross-modality information. Then based on our 
proposed new schemes, “Generalized Multiple-Instance 
Learning” and “Uncertain Labeling Density”, the system 
conjectures relevance scores of visual concepts. From these 
scores, Support Vector Regression is used to build visual 
models. Our proposed algorithm is tested on several 
concepts in large video databases. Preliminary experiments 
show promising results.


Prof. Sun's vitae:

     Ming-Ting Sun (S’79-M’81-SM’89-F’96) received the 
B.S. degree from National Taiwan University in 1976, the 
M.S, degree from University of Texas at Arlington in 1981, 
and the Ph.D. degree from University of California, Los 
Angeles in 1985, all in electrical engineering.  

     Dr. Sun joined the University of Washington in August 
1996 where he is now a Professor. Before that, he was the 
Director of the Video Signal Processing Research Group at 
Bellcore. At Bellcore, he led a group of researchers 
participated in ITU and MPEG standards meetings, and 
provided significant contributions in the development of H-
series and MPEG video coding standards.  His research 
interests include video coding and networking, multimedia 
technologies, and VLSI for signal processing.  Dr. Sun has 
been awarded 10 patents and has published more than 160 
technical papers in top journals and conferences. He has 
authored or co-authored 12 book chapters in the area of 
video and multimedia technology, and has co-edited a book 
on Compressed Video over Networks. Dr. Sun has been very 
active in professional activities and has served in various 
leadership positions including the Chair of the IEEE CAS 
Standards Committee from 1988 to 1991, the Conference Co-
Chair for the International Conference on Compression 
Technologies and Standards for Image and Video 
Communications in 1995, the Editor-in-Chief of IEEE 
Transactions on Circuits and Systems for Video Technology 
(TCSVT) from 1995 to 1997, the General Co-Chair of Visual 
Communication and Image Processing in 2000, a Distinguished 
Lecturer of the IEEE Circuits and Systems Society from 2000 
to 2001, and the Editor-in-Chief of IEEE Transactions on 
Multimedia (TMM) which is supported by four IEEE Societies, 
from 2000 to 2001. Dr. Sun has received many awards 
including an Award of Excellence from Bellcore in 1987 for 
his work on Digital Subscriber Line, an Award of Excellence 
from IEEE Standards Committee for his leadership in 
developing the IEEE IDCT Standard, the TCSVT Best Paper 
Award in 1993, and the Golden Jubilee Medal from the IEEE 
CAS Society in 2000.  Dr. Sun is a Fellow of the IEEE.