Photogrammetric point clouds offer immense potential for various applications, especially for the AEC industry and ”as-built” BIM. However, despite many advantages such as time and cost efficiency, image based point clouds of indoor environments mostly suffer from inhomogeneous and strongly fluctuating point-wise uncertainties. This lack of area-filling geometric reliability represents a strong barrier for innovations and further development of image based applications for as-built BIM, regarding both software and hardware. Therefore, this paper presents a method for the geometric verification of indoor BIMs by images and uncertainty management in order to unleash the potential of photogrammetry in context of professional building documentation heading towards ”digital twinning”. Individual 3D point accuracies, object’s surface characteristics and BIM related uncertainties according to the Level of Accuracy (LOA) specification are assessed and taken into account. The final decision of whether or not a photogrammetric point cloud confirms a given model within its associated level of accuracy results from a combined reasoning pipeline based on Dempster–Shafer evidence theory. The novel Pho-to-BIM verification method is demonstrated on three real indoor construction sites, each 3D mapped with different image sensors. Based on the experiments it is shown how to set up belief functions for evidence based reasoning individually, depending on the measurement and site characteristics.     «

Photogrammetric point clouds offer immense potential for various applications, especially for the AEC industry and ”as-built” BIM. However, despite many advantages such as time and cost efficiency, image based point clouds of indoor environments mostly suffer from inhomogeneous and strongly fluctuating point-wise uncertainties. This lack of area-filling geometric reliability represents a strong barrier for innovations and further development of image based applications for as-built BIM, regardin...     »