The H.264/AVC (Advance Video
Coding) is the latest video coding standard
established by the joint ITU-T and ISO/IEC
MPEG groups. The goals of this new
standardization were ambitious: (1) to
develop a simpler and straightforward video
coding platform; (2) to offer enhanced
compression performance at improved visual
quality, and (3) more "network-friendly"
video representation. H.264/AVC has achieved
significant improvement in the
rate-distortion efficiency, and provides,
typically, factor of two in bit-rate savings
when compared with existing MPEG-2 Video.
This short course reviews the
H.264 major sub-functional units, and
discusses the use of integer transform,
motion-compensation, deblocking filter, and
binary arithmetic coding with its use of
adaptive contexts. Life demonstration of
video encoding will be given. The software
execution workload call-graph analyses will
be reviewed for possible efficient SW/HW
implementations. Arithmetic coding,
contrasting to Huffman coding, is an
extremely powerful and yet largely unknown
coding method. This class will review and
discuss the base of Arithmetic coding in
depth. Coding efficiency of Arithmetic
coding and Huffman coding will be discussed.
The use of adaptive context and
approximated-counting state machine will be
outlined
Pi Sheng Chang, Ph.D., CISCO
System, Adjunct Prof. SCU
Deepti Sachin Joshi, SJSU
Master student
Vrinda Pariyangat, SJSU
Master student
Joe-Ming Cheng, Ph.D., Hitachi
GST, Part-time Faculty SJSU
Pi Sheng
Chang, Ph.D. (IEEE S'88-M'98-SM'04) received the B.S. degree from Federal
University of Rio de Janeiro, Rio de Janeiro, Brazil, in 1988, the M.S.
degree from Pontifical Catholic University, Rio de Janeiro, in 1990, and the
Ph.D. degree from University of California, Los Angeles, in 1998, all in
electrical engineering.
From 1994 to
1996, he was Research Assistant with the Department of Neurology,
University of California, Los Angeles, where he worked on eye movement
studies. From 1996 to 2000, he was with VTEL Corporation, Sunnyvale, CA,
working on acoustic echo cancellation and talker localization using
microphone arrays. Since 2000, he has been with Cisco Systems working on
VoIP. He is also Adjunct Professor of the Department of Computer Engineering
at Santa Clara University and Vice-Chair of IEEE Signal Processing Society,
Santa Clara Valley Chapter.
His current
interests include audio and video processing and coding, VoIP, FoIP,
adaptive signal processing and array processing with applications to
acoustics, echo cancellation, microphone arrays and spectral and DOA
estimation.
Joe-Ming
Cheng is a member of performance analysis team at Hitachi Global Storage
Technology and is also a part-time faculty member of San Jose State
University. He has worked at IBM Research and Development laboratories for
25 years on data compression algorithm and high-performance IC engine
developments. He also worked on SOC, storage system, embedded system, and
design automation tool developments. He received an Outstanding Innovation
Award and an Outstanding Achievement Award from IBM. He also worked on
guidance system and air-borne CPU developments. Dr. Cheng received an M.S.
in Scientific (Biomedical Electronics) Instrumentation, UC Santa Barbara,
1975; and a Ph.D. in Computer Engineering, UC Santa Cruz, 1996. He holds
eight issued U.S. Patents.
