Jian Lu
Vobile, Inc.
4699 Old Ironsides Dr, Suite 430
Santa Clara, CA 95054
Phone: 408-217-5010
Fax: 408-212-8300
Email: jian@computer.org


DVP term expires December 2012

Jian Lu received his BE in mechanical engineering from Zhejiang University, China,
in 1984, and PhD in electrical engineering from Dartmouth College, NH, in 1993.
From 1994 through 1996, he held research positions at University of California,
Davis, and Lawrence Livermore National Lab. He joined Apple in 1996 as a Senior
Scientist and later became a Principal Scientist. He led video and image codec
development for Apple's QuickTime technology and Final Cut professional video
products. He also served as the Principal Representative for Apple in the
development of JPEG-2000 and MPEG-4 international standards for image and video
coding. In 2005, he co-founded Vobile, Inc, where he is currently the CTO. He is the
inventor and architect of Vobile VideoDNA™ content identification technology that is
adopted by all major Hollywood movie studios and TV networks as well as content
services websites.

Dr. Lu's areas of expertise include image and video coding, content fingerprinting
and identification, media streaming and content delivery, and wavelets. He has over
40 technical publications and 7 issued U.S. patents with several more pending in
these areas. He was an invited speaker for a Special Session on Media Fingerprinting
at the 2009 SPIE/IS&T Conference on Media Forensics and Security, an organizer of
a Special Session on Content Fingerprinting and Applications at IEEE Conference on
Multimedia and Expo (ICME’09), and a keynote speaker at the IEEE Workshop on
Multimedia Security and Content Protection in 2009. He also served as a Technical
Program Committee member on several IEEE and ACM conferences.

Dr. Lu is a Senior Member of IEEE with Computer, Communications, Signal
Processing, and Consumer Electronics Societies. He has previously served as Vice
Chair and Program Chair in the Santa Clara Valley Chapter of IEEE Signal Processing
Society.


Video Fingerprinting and Its Applications
Research that began a decade ago in video copy detection has developed into a
technology known as “video fingerprinting.” Today, video fingerprinting is an
essential and enabling tool that is adopted by major Hollywood studios and TV
networks as well as content service websites for content identification and
management in online video distribution. This presentation provides a tutorial
review of video fingerprinting technology and its applications.
The presentation includes a survey on video fingerprinting algorithms to illustrate
the design aspects and tradeoffs in achieving robustness and discriminability. It
also reviews the design of fingerprint matching algorithms, including their
complexity, and the development on approximation and optimization for fast
fingerprint matching. On the application side, the presentation samples a few
commercial applications of video fingerprinting based on real-world systems and
workflows. Lastly, it offers a perspective on where the technology is headed,
including topics for continuing research and development, and trends in new
applications.

Tracking Content on the Internet
Content distribution on the Internet is marked by wide syndication. Rights holders
routinely syndicate their content to a multitude of Internet destinations to aggregate
a larger audience. Users may also participate in syndicating and proliferating
content that they may not own by sharing the content on some Internet sites. The
diverse channels of Internet content syndication and distribution make it difficult to
account for where the content is, to measure how it is consumed, and to manage the
copyrights. Content tracking is a technology that has emerged recently to provide a
solution.

This presentation gives an overview of the common architecture and workflow for
tracking different types of content including text, audio, and video. It also provides
an in-depth coverage on the design of video content tracking systems. Examples are
given to illustrate the systems that are designed for tracking pre-published and live
video content on the Internet. Experience and stories of real-world content tracking
applications are shared.