iScope: Personalized Multi-Modality Image Search for Mobile Devices

Li Shang, Department of Electrical, Computer, and Energy Engineering, University of Colorado, Boulder, CO

Abstract: Personal, portable communication and computation devices are now part of hundreds of millions of lives, often in the form of smart-phones. Emerging mobile applications and services are the main driving forces of the prevalence of personal mobile systems. Mobile devices are the first-level interface for capturing and sharing multimedia data such as images. They are therefore a natural image data management platform. Managing image data on mobile devices, however, is a challenging problem. A picture may be worth a thousand words, but without knowing the words, search can be very difficult. Manual image annotation is tedious and time consuming. Albeit the recent progress in content- based image retrieval, managing image data on personal mobile systems is challenging due to large data set size, content diversity, heterogeneous user interests and usage patterns, and resource constraints.

In this talk, Prof. Shang will present iScope, a personal content management platform. iScope is a user-centric design targeting energy-constrained distributed mobile environments. It leverages both personal context information and efficient content search techniques, as well as online learning techniques, to deliver personalized, energy-efficient content search services through a collaborative search environment, thus facilitating information discovery and social interaction.

About the speaker: Li Shang received his B.E. with honors from Tsinghua University and his PH.D. from Princeton University. He is currently an Assistant Professor in the Department of Electrical, Computer, and Energy Engineering, University of Colorado at Boulder. He has published in the areas of embedded systems, design automation, design for nanotechnologies, and computer systems. He has 59 peer-reviewed publications in these areas. Several of his work has been nominated for the Best Paper Award, including multi-scale thermal (ICCAD 2008), hybrid SET/CMOS reconfigurable architecture (DAC 2007), and thermal-aware incremental design flow (ASP-DAC 2006). His work on temperature-aware on-chip network was selected for publication in MICRO Top Picks 2006. His work also won the Best Paper Award at PDCS 2002. He is currently serving as an Associate Editor of IEEE Transactions on VLSI Systems, and the technical program committees of several top conferences on embedded systems, computer architecture, and design automation. He won his department's Best Teaching Award in 2006.