MultiMedia and the Semantic Web

Ever wanted to know how Semantics can improve and analyze Multimedia?

Dr. Gerald Friedland will discuss his work on "Advances in Semantic Analysis of Multimedia" with the International Computer Science Institute, which is associated with UC Berkeley.

We are being hosted by our good friends at Franz (https://www.franz.com) and I look forward to seeing you all there. Space is limited so sign up early!

More information about the talk is below.

Summary:

With bandwidth constantly increasing and video cameras being integrated in almost any handheld device currently on the market, the web consists more and more of multimedia data. Both "homegrown" as well as TV broadcast images and videos are a growing part of the Internet that is already taking up more space than conventional text-based media. Unfortunately, the current standard of semantic video annotation is on the level of meta data that describes the entire video in about a paragraph of text along with rudimentary meta tags (cf. YouTube or the Comcast program guide). For search and navigation in videos and for discovering content inside multimedia a more fine-granular semantic annotation is, of course, highly desired. However, the information density of perceptually encoded data makes manual semantic modeling even more difficult. Along examples of our work, this talk describes research progress and advances in automatic extraction of semantic annotation from multimedia data and how these might change the everyday Internet experience of everybody. Among other applications, I will present an online service that semi-automatically selects objects in images, a portal that automatically extracts information from Seinfeld episodes to allow browsing punchline by punchline, and a meeting recorder and browser that automatically produces a transcription of the event. While the presentation will contain some technical parts, the main focus will be to point out applicability and to discuss future directions for research and development.

Bio: (https://www.gerald-friedland.org)

Dr. Gerald Friedland is a research scientist at the International Computer Science Institute, a lab affiliated with the University of California, Berkeley. In addition to being the site manager for the EU-funded project AMIDA and the Swiss-funded IM2 project, both of which explore multimedia meeting analysis, Dr. Friedland is helping start-up Appscio.com, a Silicon Valley-based company that develops an open-source multimedia content analysis framework as a principal scientist and member of the Technical Advisory Board. He co-founded the IEEE International Conference on Semantic Computing and is a proud founder and program director of the IEEE International Summer School on Semantic Computing at UC Berkeley. He is program co-chair of the IEEE International Symposium on Multimedia 2009. He is also a program committee member of ACM Multimedia 2009 and IARIA MMedia 2009. Dr. Friedland has published more than 80 articles in conferences, journals, and books and is currently authoring a new textbook on multimedia computing. He regularly reviews for IEEE Transactions on Acoustics, Speech, and Language Processing, IEEE Transaction on Multimedia, IEEE Multimedia, Springer’s Machine Vision and Application, and other journals. He is the recipient of several research and industry recognitions, among them the European Academic Software Award and the Multimedia Entrepreneur Award by the German Federal Department of Economics. Dr. Friedland received his doctorate (summa cum laude) and master’s degree in computer science from Freie Universitaet Berlin, Germany, in 2002 and 2006, respectively.