Harnessing Indoor 3D Point Cloud Reconstruction for Automated Scan-to-BIM Workflows: A Systematic Review

The rapid growth of urban environments has increased the demand for digital representations of indoor spaces that enable automation and intelligent management. Building Information Modeling (BIM) plays a key role in this transformation by converting raw point cloud data into accurate, semantically rich models for applications such as smart building technologies, facility management, predictive maintenance, disaster preparedness, and urban digitalization. While scan-to-BIM workflows have been extensively studied, most existing reviews focus on individual stages or technologies rather than providing a comprehensive, end-to-end perspective. This paper addresses this gap by offering a systematic review of the full scan-to-BIM modeling process. It examines current approaches for reconstructing structural components, covering both traditional geometric methods and emerging deep learning techniques. The review also highlights key advances and challenges related to automation, data quality, object complexity, and workflow standardization. Overall, this work aims to improve scan-to-BIM practices and support the development of more reliable and efficient BIM models for indoor environments.