PhD Defense: "Design and Calibration of Wide-Area Camera Networks"

Thomas Kuo

August 21st (Wednesday), 1:00pm
Harold Frank Hall (HFH), Rm 4164

Advancements in sensing, processing, communications, and power technologies have enabled vast numbers of inexpensive wireless cameras that can observe a wide area. This dissertation addresses the challenges of designing a wide-area camera network and finding the spatial and temporal relationships between the cameras. Much of this dissertation is motivated by the Scalable, Large Optical Sensor Netowrk (SCALLOPSNet), a wireless, wide-area camera network deployed across the UCSB campus.

The first challenge this work addresses is the calibration of non-overlapping cameras into a geographic coordinate system using mobile devices. Then we explore the problem of the data-driven alignment and synchronization of videos that have different start times and variable frame rates. Our solution involves finding and matching the intersections of object trajectories, and, then, using them as control points for alignment and synchronization.

About Thomas Kuo:

Thomas Kuo received his B.S. in Electrical Engineering and Computer Science in 2004 from UC Berkeley and his M.S. from UC Santa Barbara in 2006. He has been working with in the Vision Research Lab at UCSB to deploy the SCALLOPSNet camera network and is interested in research in the calibration of camera networks, and has interned at HRL, researching person detection. After graduation, he plans to work at a startup, EcoLek, develop Birdeez and other apps to help connect people to nature.

Hosted by: Professor B.S. Manjunath