期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2008
卷号：XXXVII Part B3a
页码：259-264
出版社：Copernicus Publications
摘要：We present a 3D building reconstruction method from satellite images based on a stochastic approach. It consists in reconstructing buildings by assembling simple urban structures extracted from a library of 3D parametric models, as a LEGO R game. Such a method is particularly well adapted to data of average quality such as high resolution satellite images. The approach is based on a density formulation defined within a Bayesian framework. The configuration which maximizes this density is found using a RJMCMC sampler which is efficient w.r.t. the multiple parametric object recognition problem. Experimental results are shown on complex buildings and dense urban areas using PLEIADES simulations
