Deepak S. Turaga
Automating Large-Scale Multimedia Exploration for Smarter Planet Applications
Deepak Srinivas Turaga received the B.Tech. degree in Electrical Engineering from Indian Institute of Technology, Bombay in 1997 and the M.S. and Ph.D. degrees in Electrical and Computer Engineering from Carnegie Mellon University, Pittsburgh in 1999 and 2001, respectively. He is currently a Manager of the Exploratory Stream Analytics department at the IBM T. J. Watson Research Center in Hawthorne. He was at Philips Research during 2001-2003 and with Sony Electronics in 2003-2004.
His research interests lie primarily in stream mining, statistical signal processing, multimedia coding and streaming, machine learning and computer vision applications. In these areas he has published over 50 journal and conference papers and one book and two book chapters. He has also filed over twenty invention disclosures, and has participated in MPEG standardization activities.
He received the CSVT 2006 Transactions Best Paper Award (with M. van der Schaar and B. Pesquet-Popescu), and is a coauthor for the 2006 IEEE ICASSP Best Student Paper (with H. Tseng, O. Verscheure and U. Chaudhari). He is an Associate Editor of the IEEE Transactions on Circuits and Systems for Video Technology, and an Associate Editor for the Hindawi Journal for the Advances in Multimedia. He was an Associate Editor of the IEEE Transactions on Multimedia during the period 2006-08. He is also currently an Adjunct Associate Professor in the Electrical Engineering Department at Columbia University.
Content Delivery Considerations for Different Types of Internet Video
This talk will look at the issues with delivering a great video experience on behalf of many Google properties with different constraints. We will primarily focus on on-demand video as it comprises the bulk of video demand, but will also look at the constraints and needs imposed by live video and video conferencing. An overview of the delivery architecture and demand statistics will be covered as well.
Leonidas Kontothanassis is currently Engineering Director at Google, Inc. in Cambridge, Massachusetts with a focus on Network Infrastructure and Web Acceleration Technologies. Leonidas has spent significant time both as a researcher with DEC, Compaq and HP Labs, as well as a lead engineer with Akamai Inc, and Intel Inc. in the field of network optimization. He has also worked in such varied fields as computer architecture, parallel programming, scheduling, and medical device analysis and instrumentation. Since joining Google in October 2006, the office has grown from 10 to over 350 people, with engineering teams working on Book Search, Image Search, Chrome, ChromeOS, YouTube, and networking infrastructure many of them under his direct supervision. Leonidas also held an adjunct faculty position with Boston University between 1997 and 2001. He is the author or co-author of dozens of academic papers and patents and a member of the ACM. Leonidas received his B.S. in Computer Engineering from the University of Patras, Greece in 1990 and his M.S. and Ph.D. degrees in Computer Science from the University of Rochester in 1992, and 1996 respectively.
Jim van Welzen
Challenges Leveraging the Burgeoning Power of Mobile Multimedia Architectures for Innovative Commercial Multimedia Applications
Mobile architectures targeted at today’s smartphones and tablet devices offer an increasingly powerful combination of video, camera, compute, and graphics capabilities which enable a new array of sophisticated multimedia applications. Developers are migrating high definition video conferencing, premium content delivery, and video editing software, once restricted to the domain of tethered set top boxes and personal computers , to commercial mobile platforms. Furthermore, the inherent mobility of these devices and usage model (they go where we go and can see what we see) coupled with its processing horsepower creates the opportunity for rich camera solutions leveraging computational photography and innovative, environment aware, mixed media interfaces leveraging computer vision and augmented reality. In practice, however, enabling even a moderately sized software vendor to implement an advanced multimedia application on powerful mobile architecture represents a herculean challenge for all members of the ecosystem: those that provide the silicon, operating system, middleware, and the application itself. It requires designing a multimedia system with an intelligent, use case cognizant division of labor between the architecture’s components cast in the native frameworks of the operating system, optimizing internal data and control paths between those components, and tempering application interfaces with both the appropriate amount of flexibility and abstraction. This talk discusses both these challenges of delivering sophisticated mobile multimedia applications and the approaches used in practice to surmount them.
Jim van Welzen is the Director of Multimedia Software for NVIDIA’s Mobile business unit where he leads an international team in delivering multimedia solutions into smartphones, tablets, and other mobile devices. His disciplines include multimedia system software architecture, performance and power optimizations, and application enablement. Jim is active in the Khronos international standards group particularly the OpenMAX multimedia standards where he has served both as chair and specification editor. He holds several patents in multimedia and graphics and is published in IEEE proceedings. Jim received his Masters degree in Computer Science from the University of North Carolina, Chapel Hill and his Bachelors from North Carolina State University.