This page is provided for historical and archival purposes only. While the seminar dates are correct, we offer no guarantee of informational accuracy or link validity. Contact information for the speakers, hosts and seminar committee are certainly out of date.
Digital Equipment Corporation Cambridge Research Lab
Computer graphics and virtual reality are making rapid progress in displaying 3D scenes of ever increasing complexity and realism. Unfortunately, relatively little attention is paid to the preliminary stage of model building, which can consume most of the time in any large animation or virtual reality project. In this talk, I will examine automated techniques for extracting 3D scenes from multiple views and/or video sequences. In particular, I will address the simpler problem of compositing multiple pieces of a planar scene (e.g., a whiteboard, a document, a bookshelf) into a larger high-resolution image. I will also discuss the more difficult general 3D scene recovery problem. Potential applications of these techniques include low-cost portable scanners (the whiteboard example), architectural walk-throughs of existing (e.g., historic) buildings, and a new home-shopping paradigm where you browse your familiar supermarket aisles.
Host: Yangsheng Xu (firstname.lastname@example.org) Appointment: Ava Cruse (email@example.com)
Richard Szeliski received the B. Eng. degree in Honours Electrical Engineering from McGill University, Montreal, the M. A. Sc. degree in Electrical Engineering from the University of British Columbia, Vancouver, and the Ph. D. degree in Computer Science from Carnegie Mellon University, Pittsburgh (in 1988).
After graduating from CMU, Richard worked at Schlumberger Palo Alto Research, in Palo Alto, and at the Artificial Intelligence Center of SRI International, Menlo Park. Since July 1989 he has been a Member of Research Staff at the Cambridge Research Lab of Digital Equipment orporation, Cambridge, where he is pursuing research in 3-D computer vision, geometric modeling, medical image registration, and parallel programming and algorithms.