Streamlined single software workflow with ML-based point cloud clustering for virtual reality building inspection

1 . A method comprising:

storing, by a computing device executing a single software workflow, an as-planned building information model (BIM) data for a building project, wherein the as-planned BIM data comprises a 3D design model of the building project;

acquiring, by the computing device executing the single software workflow, an as-built point cloud data of the building project;

processing, via machine learning by the computing device executing the single software workflow, the as-built point cloud data to generate cluster points;

generating, via machine learning by the computing device executing the single software workflow, 3D geometry and mesh data for each cluster point set that correspond to different building components;

converting, by the computing device executing the single software workflow, the 3D geometry and mesh data into as-built BIM data comprising an as-built 3D model, in a BIM file format, constructed from the different building components;

comparing, by the computing device executing the single software workflow, the as-built BIM data and the as-planned BIM data and determining a discrepancy between geometries of the as-built BIM data and the as-planned BIM data;

outputting, by the computing device executing the single software workflow, an inspection report identifying the discrepancy between geometries of the as-built point cloud data and the as-planned BIM data;

performing, by the computing device executing the single software workflow, virtual reality visualization of the as-built 3D model comprising the 3D geometries generated from the as-built point cloud data and correlating the 3D geometries with the as-planned BIM data and showing the discrepancy between the geometries of the as-built BIM data and the as-planned BIM data; and

performing, by the computing device executing the single software workflow, compliance checking of the as-built 3D model by comparing the as-built 3D model against building rules and codes.

2 . The method of claim 1 , wherein the cluster points are generated using a density-based spatial clustering of applications with a noise (DBSCAN) algorithm performed by the computing device.

3 . The method of claim 1 , wherein the as-planned BIM data is imported from an Autodesk Revit file.

4 . The method of claim 1 , wherein the as-built point cloud data is acquired from an imaging device or camera using LiDAR technology.

5 . The method of claim 1 , wherein an as-planned 3D model is overlayed on top of the as-built 3D model of the building project, wherein the as-planned BIM data comprises the as-planned 3D model.

6 . The method of claim 1 , wherein the as-built point cloud data comprises thermal imaging data and/or LiDAR data and/or digital laser scanned and/or digital photographs of the building project.

7 . A system comprising:

at least one processor; and

memory configured to communicate with the at least one processor, wherein the memory stores instructions that, in response to execution by the at least one processor, cause the at least one processor to perform operations within a single software workflow comprising:

storing an as-planned building information model (BIM) data for a building project, wherein the as-planned BIM data comprises a 3D design model of the building project;

acquiring an as-built point cloud data of the building project;

processing, via machine learning, the as-built point cloud data to generate cluster points;

generating, via machine learning, a 3D geometry and mesh data for each cluster point that correspond to different building components;

converting the 3D geometry and mesh data into as-built BIM data comprising an as-built 3D model, in a BIM file format, constructed from the different building components;

comparing the as-built BIM data and the as-planned BIM data and determining a discrepancy between geometries of the as-built BIM data and the as-planned BIM data;

outputting an inspection report identifying the discrepancy between geometries of the as-built point cloud data and the as-planned BIM data;

performing virtual reality visualization of the as-built 3D model comprising the 3D geometries generated from the as-built point cloud data and correlating the 3D geometries with the as-planned BIM data; and

performing compliance checking of the as-built 3D model with building rules and codes.

8 . The system of claim 7 , wherein the cluster points are generated using a density-based spatial clustering of applications with noise (DBSCAN) algorithm performed by the at least one processor.

9 . The system of claim 7 , wherein an as-planned 3D model is overlayed on top of the as-built 3D model of the building project, wherein the as-planned BIM data comprises the as-planned 3D model.

10 . The system of claim 7 , wherein the as-built point cloud data is acquired from an imaging device or camera using LiDAR technology.

11 . The system of claim 7 , wherein the as-built point cloud data comprises thermal imaging data and/or LiDAR data and/or digital laser scanned and/or digital photographs of the building project.

12 . A non-transitory, tangible computer-readable storage medium having instructions stored thereon that, in response to execution by a computer-based system, cause the computer-based system to perform operations within a single software workflow comprising:

storing an as-planned building information model (BIM) data for a building project, wherein the as-planned BIM data comprises a 3D design model of the building project;

acquiring an as-built point cloud data of the building project;

processing, via machine learning, the as-built point cloud data to generate cluster points;

generating, via machine learning, a 3D geometry and mesh data for each cluster point that correspond to different building components;

converting the 3D geometry and mesh data into as-built BIM data comprising an as-built 3D model, in a BIM file format, constructed from the different building components;

comparing the as-built BIM data and the as-planned BIM data and determining a discrepancy between geometries of the as-built BIM data and the as-planned BIM data;

outputting an inspection report identifying the discrepancy between geometries of the as-built point cloud data and the as-planned BIM data;

performing virtual reality visualization of the as-built 3D model comprising the 3D geometries generated from the as-built point cloud data and correlating the 3D geometries with the as-planned BIM data; and

performing compliance checking of the as-built 3D model with building rules and codes.

13 . The non-transitory, tangible computer-readable storage medium of claim 12 , wherein the cluster points are generated using a density-based spatial clustering of applications with noise (DBSCAN) algorithm performed by the computer-based system.

14 . The non-transitory, tangible computer-readable storage medium of claim 12 , wherein an as-planned 3D model is overlayed on top of the as-built 3D model of the building project, wherein the as-planned BIM data comprises the as-planned 3D model.

15 . The non-transitory, tangible computer-readable storage medium of claim 12 , wherein the as-built point cloud data comprises thermal imaging data and/or LiDAR data and/or digital laser scanned and/or digital photographs of the building project.