3-D Reconstruction from Images

Prof. Andrea Fusiello

Computer Science

University of Verona

Date: Friday, June 9, 2006
Place: Engineering Sciences Building, 1001
Time: 2:00 pm — 3:00 pm

Abstract:

In this talk we demonstrate new methods for accurate 3D reconstruction of a scene from images with guaranteed accuracy bounds. This takes into account the propagation of data errors and roundoff. Image points are represented as small rectangles: As a result, the output of the n-views triangulation is not a single point in space, but a polyhedron that contains all the possible solutions. Interval Analysis is used to bound this polyhedron with a box. Geometrical constraints such as orthogonality, parallelism, and coplanarity are subsequently enforced in order to reduce the size of those boxes. Experiments with real calibrated images illustrate the effectiveness of the approach.

ANDREA FUSIELLO received his Laurea (Master) degree in Computer Science from the University of Udine in 1994. He received the Dottorato di Ricerca (PhD) in Computer Engineering from the University of Trieste in 1999. He was a Visiting Research Fellow at Heriot-Watt University, Edinburgh in 1999. From 2001 to 2004 he served as a Ricercatore (Assistant Professor) at the Department of Computer Science, University of Verona. He is now Professore Associato (Associate Professor) at the same department, where he teaches Computer Vision and Computer Graphics. He has been advisor of more than 20 master students and 3 PhD students.

His research is mainly focused on Computer Vision (image analysis, model acquisition), with applications to Human-Computer Interaction (cross-modality) and Computer Graphics (image based rendering, augmented reality). He published more than 50 papers in the field. Andrea Fusiello is a member of the IAPR, EUROGRAPHICS and IEEE CS.

.

Host: B.S. Manjunath, Professor of ECE