This paper presents a knowledge-based approach for automatic 3D reconstruction of buildings from aerial images. The building reconstruction process is described as a tree search in the space of possible building hypotheses. Hypotheses derived from outlines of building footprints from the map are fit against image pixel gradients. To guide the search of the tree an evaluation function based on information theory principles is defined. The proposed evaluation function defines the score of matching between a hypothesised building model and the image pixel gradients.
Home Contents Author index Keyword index
This document produced for BMVC 2001