Abstract. In this paper we present an eicient modeling framework for large scale urban scenes. Taking surface meshes derived from multiview-stereo systems as input, our algorithm outputs simpliied models with semantics at diferent levels of detail (LODs). Our key observation is that urban building is usually composed of planar roof tops connected with vertical walls. There are two major steps in our framework: segmentation and building modeling. The scene is irst segmented into four classes with a Markov random ield combining height and image features. In the following modeling step, various 2D line segments sketching the roof boundaries are detected and slice the plane into faces. Through assigning each face with a roof plane, the inal model is constructed by extruding the faces to the corresponding planes. By combining geometric and appearance cues together, the proposed method is robust and fast compared to the state-of-the-art algorithms.