Energy Conservation for Image Retrieval on Mobile Systems

Title: Energy Conservation for Image Retrieval on Mobile Systems
 
Time:10:30am, Oct 26th, 2011

Where:Room 200, No.3 Seiee Building

Abstract:

Mobile systems have become major producers of multimedia contents. A smart phone can store thousands of images and this creates the need for better organization and retrieval of images. Content-Based Image Retrieval (CBIR) is a method to retrieve images based on their visual contents, instead of using tags entered by users. CBIR is computation and memory intensive and consumes significant amounts of energy. This study examines energy conservation for CBIR on mobile systems. We present three improvements to save energy: selectively loading image features, adaptively loading features based on overall similarities, and caching features in memory. We further investigate if energy can be saved by migrating parts of the computation to a server. We consider several factors, including wireless data rate, server speed, number of indexed images, and the number of image queries.

 

Biography: 

Yung-Hsiang Lu is a visiting associate professor in the Department of Computer Science of National University of Singapore in Fall 2011. He is an associate professor in the School of Electrical and Computer Engineering of Purdue University, U.S.A. He received Ph.D. from Stanford University in 2002. His research focuses on energy conservation and resource management in computer systems, embedded systems, sensor networks, and mobile robots. He received a Career Award from the National Science Foundation in 2004 and the Purdue Class 1922 Helping Student Learn Award in 2008.

 
Dr. Lu is a senior member of the IEEE and the ACM. He is an associate editor for ACM Transactions on Embedded Computing Systems and ACM Transactions on Design Automation of Electronic Systems. He is chair of Green Multimedia Interest Group in the IEEE Multimedia Communication Technical Committee and vice chair of ACM SIGDA Low Power Technical Committee. His paper "Quantitative Comparison of Power Management Algorithms" is selected in the book "Design, Automation, and Test in Europe: The Most Influential Papers of 10 Years DATE", Springer, ISBN 9048176530.
[ 2011-10-25 ]