Abstract:Large volumes of 3D parametric datasets, such as building information modeling (BIM), are the foundation for developing and applying smart city and digital twin technologies. Those datasets are also considered valuable tools for efficiently managing rebuilt structures during the operation and maintenance stages. Nevertheless, current approaches developed for the scan-to-BIM process rely on manual or semi-automatic procedures and insufficiently leverage semantic data in point clouds. These methods struggle to accurately represent large-scale indoor complex layouts and extract details from irregular-shaped unstructured elements, causing inefficiencies in BIM model generation. To address these issues, we propose an innovative scan-to-BIM framework based on deep learning algorithms and raw point cloud data, enabling the automatic generation of 3D models for both structured and unstructured indoor elements. Initially, we propose an enhanced deep learning neural network to improve the point clouds' semantic segmentation accuracy. Subsequently, an efficient workflow is developed to reconstruct 3D building models of structured indoor scenes. The proposed workflow can reconstruct large-scale data with multiple room layouts of Manhattan or non-Manhattan structures and reconstruct 3D models automatically by using a BIM parametric algorithm implemented in Revit software. Moreover, we introduce a robust method for unstructured elements to automatically generate corresponding 3D BIM models, even when the incorporating semantic information is incomplete. The proposed approach was evaluated on synthetic and real data for different scales and complexities of indoor scenes. The results of the experiments demonstrate that the improved model significantly enhances the overall semantic segmentation accuracy compared to the baseline models. The proposed scan-to-BIM framework is efficient for indoor element 3D reconstruction, achieving precision, recall, and F-score values ranging from 96% to 99%. The generated BIM models are competitive with traditional methods regarding model completeness and geometric accuracy.

From dense point clouds to semantic digital models: End-to-end AI-based automation procedure for Manhattan-world structures

Automated BIM generation for large-scale indoor complex environments based on deep learning

An Automatic Building Models' Parametrer Reconstruction Method from Point Clouds

A Specialized Pipeline for Efficient and Reliable 3D Semantic Model Reconstruction of Buildings from Indoor Point Clouds

Automatic Reconstruction of Semantic 3D Models from 2D Floor Plans

Towards Automatic Generation of As-Built BIM: 3D Building Facade Modeling and Material Recognition from Images

Leveraging Deep Learning for Automated Reconstruction of Indoor Unstructured Elements in Scan-to-BIM

3D Point Cloud Semantic Modelling: Integrated Framework for Indoor Spaces and Furniture

Automated Semantic Segmentation of Indoor Point Clouds from Close-Range Images with Three-Dimensional Deep Learning

Combining visibility analysis and deep learning for refinement of semantic 3D building models by conflict classification

Enhancing point cloud semantic segmentation in the data‐scarce domain of industrial plants through synthetic data

A Hybrid Semantic-Geometric Approach for Clutter-Resistant Floorplan Generation from Building Point Clouds

Automatic 2D Floorplan CAD Generation from 3D Point Clouds

A Hybrid Approach to Semi-Automatic Conversion of 3D Point Cloud Data to Building Information Model

AUTOMATIC 3D BUILDING MODEL GENERATION USING DEEP LEARNING METHODS BASED ON CITYJSON AND 2D FLOOR PLANS

A Comprehensive Automated 3D Approach for Building Extraction, Reconstruction, and Regularization from Airborne Laser Scanning Point Clouds

Automatic Extraction and Reconstruction of a 3D Wireframe of an Indoor Scene from Semantic Point Clouds

Semantic-aware Room-Level Indoor Modeling from Point Clouds

Automatic Structural Mapping And Semantic Optimization From Indoor Point Clouds

APC2Mesh: Bridging the gap from occluded building façades to full 3D models

Enriching geometric digital twins of buildings with small objects by fusing laser scanning and AI-based image recognition