Automatic Floor Matching for 3D Indoor Spatial Modeling

Abstract

With the advent of a variety of indoor location-based services, the necessity of 3D indoor model construction has become a significant issue worth noting and following. The aim of this study is to propose an algorithm of floor matching to construct a multi-floor building model. In this case, the characteristics of shape and position of lift features are used to search matched pairs in the algorithm. In addition, the vertical connectivity information also can be generated through the process. The proposed algorithm was applied to the Seoul National University Library to verify its suitability. In the case of a high-rise building, it is expected that a multi-floor building model can be constructed efficiently by automatically aligning the data generated per each floor through the method developed in this study.