Seminars/Colloquia/Invited Talks

Seminars

David Kriegman

Beyond Lambert: Techniques for Reconstructing Surface Shape

PLACE:	Clark 314
EVENT:	CIS Seminar Series
DATE:	September 20, 2005
TIME:	1:00 - 2:00

Abstract

We address the problem of reconstructing the shape of objects with non-Lambertian reflectance (i.e., with an arbitrary and possibly anisotropic and spatially varying bidirectional reflectance distribution function (BRDF)). Here, we present three solutions to this problem that were developed by re-examining the underlying image formation process; the methods make few assumptions about the object's shape, the presence or absence of shadowing, or the nature of the BRDF which may vary over the surface.
The first method takes advantage of Helmholtz reciprocity, the second method takes advantage of the dichromatic model of reflectance, and the third method exploits the fact that the radiance along a ray of light is constant. In particular, the first method uses stereo pairs of images in which point light sources are co-located at the centers of projection of the stereo cameras. The second method introduces a new color space (SUV) in which two channels are independent of the specular reflectance; using these channels, photometric stereo can be used to estimate shape. The third method is based on double covering a scene's incident light field; the depths of surface points are estimated using a large collection of images in which the viewpoint remains fixed and a point light source illuminates the object. Concurrently, an "effective BRDF" for each reconstructed surface point is obtained and can be used for rendering the object.

Brief biography

David J. Kriegman received a B.S.E. in Electrical Engineering and Computer Science at Princeton University in 1983, and was awarded the Charles Ira Young Medal for Electrical Engineering Research. He received an M.S. in 1984 and Ph.D. in 1989 in Electrical Engineering from Stanford University. From 1990-1998, he was on the faculty of the Electrical Engineering and Computer Science Departments at Yale University and from 1998-2002 he served as Associate Professor at the Computer Science Department and Beckman Institute at the University of Illinois. In 2002, he joined the Computer Science & Engineering Department at the University of California, San Diego as a Professor.
Kriegman was chosen for a National Science Foundation Young Investigator Award in 1992, and has received best paper awards at the 1996 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) and the 1998 European Conference on Computer Vision. He has served as Program Co-chair of CVPR 2000, is the Associate Editor-in-Chief of the IEEE Transactions on Pattern Analysis and Machine Intelligence, and is currently an Associate Editor of the IEEE Transactions on Robotics and Automation. He has published over one hundred papers on object recognition, illumination modeling, face recognition, structure from motion, geometry of curves and surfaces, mobile robot navigation, and robot planning.