Events Calendar

CS Colloq: Nirupama Bulusu

Time: 11 AM - 12:30 PMLocation: SAL 222Talk title: Resource-efficient Audiovisual Sensing
Speaker: Prof. Nirupama Bulusu(Portland State University)
Host: Prof. Ramesh GovindanAbstract:
Animal calls, street signs, car honks, etc. The physical world is permeated with sounds and images that can be captured and analyzed using audiovisual sensor networks. Both sensing modalities feature high sampling rates or large sample sizes making them challenging to implement on embedded platforms with sharply limited energy, bandwidth and processing resources.In this talk, I will first show how we have built resource-efficient audiovisual sensing applications on embedded platforms. Application-specific system architectures and compressive sensing algorithms reduce the amount of data sampled and transmitted at each sensor. I will present three real world case studies: cane-toad monitoring; PetrolWatch, a fuel price collection
application; and noise pollution monitoring.Second, many image sensing applications require accurate camera position and orientation in 3D space. Currently, the most accurate camera localization method requires the resource-intensive task of detecting and matching several feature points visible in frames taken from each camera, which is unreasonable for resource constrained sensors. I will describe the
design and evaluation of a highly accurate, robust, efficient, distributed 3D localization system suitable for embedded image sensors that significantly reduces the costs of the point correspondence problem, by using a special 3D target.This talk describes joint work with colleagues at Portland State University, and the University of New South Wales.Bio:
Nirupama Bulusu received the B.Tech degree from Indian Institute of Technology (IIT), Madras, India in 1997, the M.S degree from the University of Southern California in 2000 and the Ph.D degree from the University of California Los Angeles (UCLA) in 2002, all in computer science. Since 2004, she has been an Assistant Professor of Computer Science at Portland State
University. Her research interests lie in sensor networks, with an emphasis on environmental and urban sensing applications. She is a recipient of the NSF CAREER award.

Location: Henry Salvatori Computer Science Center (SAL) - 222

Audiences: Everyone Is Invited

Contact: CS Front Desk