Cumulative Assessment for Urban 3D Modeling
This addresses the need for standardized evaluation in urban 3D modeling for researchers and practitioners, though it is incremental as it builds on existing methods.
The paper tackles the problem of evaluating urban 3D modeling from satellite images by introducing a cumulative assessment metric that captures errors from semantic segmentation, 3D reconstruction, and model fitting, and provides public datasets and an end-to-end baseline solution to stimulate research.
Urban 3D modeling from satellite images requires accurate semantic segmentation to delineate urban features, multiple view stereo for 3D reconstruction of surface heights, and 3D model fitting to produce compact models with accurate surface slopes. In this work, we present a cumulative assessment metric that succinctly captures error contributions from each of these components. We demonstrate our approach by providing challenging public datasets and extending two open source projects to provide an end-to-end 3D modeling baseline solution to stimulate further research and evaluation with a public leaderboard.