A multiple view stereo benchmark for satellite imagery

The availability of public multiple view stereo benchmark datasets has been instrumental in enabling research to advance the state of the art in the field and to apply and customize methods to real-world problems. Until now, no public multiple view stereo benchmark dataset has been available for satellite imaging applications. In this work, we describe a public benchmark dataset for multiple view stereo applied to three-dimensional outdoor scene mapping using commercial satellite imagery. This dataset includes fifty Digital Globe WorldView-3 panchromatic and multispectral images of a 100 square kilometer area near San Fernando, Argentina. We also provide high-resolution airborne lidar ground truth data for a 20 square kilometer subset of this area and performance analysis software to assess accuracy and completeness metrics. We report initial results from available solutions using this benchmark data and encourage continued research by making this benchmark dataset publicly available to the research community.