Abstract
Abstract. In this paper we deal with the automatic reconstruction of interior walls from point clouds, an active research topic with several practical applications. We propose an improved version of the method presented in (Magri and Fusiello, 2018), where the overall structure of the environment is extracted by fitting lines to the main building features, using Min-hashed J-Linkage as a multi-model fitting technique. Our variation has the merit of producing more accurate results, both in terms of wall reconstruction and room segmentation, and greatly reducing the need for user-defined thresholds.
Highlights
Nowadays we are witnessing an increasing in the demand of updated and detailed 3D models of indoor environments, that find application in several fields such as navigation, emergency response, building maintenance and monitoring (Zlatanova et al, 2013)
Among the methods proposed in the literature, we focus on the one presented in (Magri and Fusiello, 2018), that is able to automatically reconstruct the interior wall of a building, relying on a multi-model fitting technique known as J-Linkage (Toldo and Fusiello, 2008)
The method operates on the 2D floor plane, where 3D points are projected, and extracts the overall structure of the environment by fitting line segments to the main building features
Summary
Nowadays we are witnessing an increasing in the demand of updated and detailed 3D models of indoor environments, that find application in several fields such as navigation, emergency response, building maintenance and monitoring (Zlatanova et al, 2013). The growing availability of instruments such as terrestrial or mobile laser scanners allows to obtain high quality 3D data, that represent the basis for the as-built indoor models reconstruction. For these reasons, generating 3D indoor models from point clouds has become an active research topic in recent years, leading to the development of algorithms that are able to reduce time and cost of the manual processing (Wang et al, 2017). Among the methods proposed in the literature, we focus on the one presented in (Magri and Fusiello, 2018), that is able to automatically reconstruct the interior wall of a building, relying on a multi-model fitting technique known as J-Linkage (Toldo and Fusiello, 2008). The aim of this work is to propose an improved version of (Magri and Fusiello, 2018), taking into account the points distribution along the vertical axis, avoiding the usage of several manually tuned thresholds and leveraging on a cell-complex subdivision of the plane to enhance the topological correctness of the result
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
