Personal tools
Home Events Events Archive 2013 Real-Time Compressive Video Using Coded Aperture Cameras

Real-Time Compressive Video Using Coded Aperture Cameras

— filed under:

What
  • Seminar Series
When Jan 14, 2013
from 01:00 PM to 02:00 PM
Where Engr. IV Bldg., Shannon Room 54-134
Contact Name
Add event to calendar vCal
iCal

Thomas Goldstein

Rice University

 
 Abstract

"Single-pixel cameras" are a new modality for image acquisition using compressive measurements.  In contrast to conventional imaging, which relies on arrays of many photodetectors, single picture cameras obtain high resolution images using only a single detector.  This is accomplished using a "coded aperture," which compresses global image information into each measurement rather than observing one pixel at a time.  In this talk, we discuss the challenges of applying these new compressive imaging tools to video acquisition.  Compressive video poses many challenges that imaging does not.  We are no longer imaging a static scene, but rather we must explicitly model the motion of objects and exploit correlations between adjacent frames.   Furthermore, real-time reconstruction of video from compressive measurements requires the use of sophisticated numerical algorithms and parallel hardware implementations.

 

Biography

Tom Goldstein is Post-Doctoral Fellow at the Rice University Department of Electrical Engineering, working with the signal processing group.  He completed his PhD at UCLA in 2010, and a post-doctoral fellowship at Stanford University in 2012.  Tom's research interests lie in large-scale optimization and its application to image processing and biological modeling. In recent years, he has focused on improved reconstruction models for computed tomography images, and real-time video reconstruction algorithms for coded aperture (i.e. compressive) cameras.

Document Actions