THREE-DIMENSIONAL BUILDING RECONSTRUCTION USING IMAGES OBTAINED BY UNMANNED AERIAL VEHICLES
- Computer Vision and Remote Sensing Berlin University of Technology Sekr. FR3-1, Franklinstr. 28/29, D-10587, Berlin, Germany
Keywords: UAVs, Photogrammetry, 3D Reconstruction, Sensor Orientation
Abstract. Unmanned Aerial Vehicles (UAVs) offer several new possibilities in a wide range of applications. One example is the 3D reconstruction of buildings. In former times this was either restricted by earthbound vehicles to the reconstruction of facades or by air-borne sensors to generate only very coarse building models. This paper describes an approach for fully automatic image-based 3D reconstruction of buildings using UAVs. UAVs are able to observe the whole 3D scene and to capture images of the object of interest from completely different perspectives. The platform used by this work is a Falcon 8 octocopter from Ascending Technologies. A slightly modiﬁed high-resolution consumer camera serves as sensor for data acquisition. The ﬁnal 3D reconstruction is computed ofﬂine after image acquisition and follows a reconstruction process originally developed for image sequences obtained by earthbound vehicles. The per- formance of the described method is evaluated on benchmark datasets showing that the achieved accuracy is high and even comparable with Light Detection and Ranging (LIDAR). Additionally, the results of the application of the complete processing-chain starting at image acquisition and ending in a dense surface-mesh are presented and discussed.