Talk: Berthold K.P. Horn on "Dynamic Reconstruction"
Talk: Berthold K.P. Horn on "Dynamic Reconstruction"
Wednesday, November 18, 2009 | 5:00pm - 6:00pm
Location: 
Roth Room (E15-283A), MIT Media Lab
Speaker: 
Berthold K.P. Horn
Host/Chair: 
Ramesh Raskar
"Dynamic Reconstruction" is the term for computational imaging methods that generalize CT, MRI and CA and provide for imaging using moving detectors and moving radiation-blocking elements. Applications include detecting IEDs at a distance using back-scattered X-rays, detecting SNM at a distance by imaging gamma rays, CA imaging with non-ideal apertures, and a new approach to cone-beam CT. If there is time, we can also discuss some of the interesting number theory related to CA and how it leads to a consideration of the eigenvectors of the DFT. Also, a demonstration that any function can be decomposed into four uniquely defined components with special FT properties (a generalization of the notion of odd and even components).
Biography: 
EECS faculty member Berthold K.P. Horn, professor of computer science and engineering and principal investigator in the Computer Science and Artificial Intelligence Lab (CSAIL), is the second recipient, since 2007, of the IEEE Computer Society's Azriel Rosenfeld Life Time Achievement Award, presented at the 12th annual International Conference on Computer Vision (ICCV) in Kyoto, Japan. Th Rosenfeld Award, sponsored by OMRON, was established in memory of computer scientist and mathematician Prof. Azriel Rosenfeld, to honor outstanding researchers who are recognized for making significant contributions to the field of Computer Vision over longtime careers. Berthold was recognized for pioneering work on early vision including optical flow and shape from shading.