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

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.

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.