Robert Collins, Chris Jaynes, Yong-Qing Cheng, Xiaoguang Wang,
Frank Stolle, Howard Schultz, Allen Hanson, and Edward Riseman,
"The UMass Ascender System for 3D Site Model Construction,"
RADIUS: Image Understanding for Imagery Intelligence,
O.Firschein and T.Strat, eds, Morgan Kaufmann, 1996, pp. 209-222.
The Automated Site Construction, Extension, Detection and Refinement
system (Ascender) has been developed to automatically populate a
site model with 3D buildings extracted from multiple, overlapping views.
Image Understanding (IU) algorithms hypothesize potential building
roofs in one image, automatically locate supporting geometric
evidence in other images, and determine the precise shape and position
of the new buildings via multi-image triangulation. Backprojecting
image intensities onto the recovered object surfaces leads to
realistic graphical site model displays, and to extraction of
symbolic features such as windows and doors. This paper describes how
the Ascender system acquires and extends an initial site model. System
performance is evaluated using imagery from Fort Hood, Texas.
Click here for
full paper (1185062 bytes, compressed using gzip).