Matterport - Procedure with Revit and AutoCAD

[Anonymous]

Hello I was wondering if someone could briefly explain to me how to work with Matterport (MP) point cloud in Revit, to achieve some goals. 

I saw some video where they use RECAP to process the MP 3D and texture, then it's possible to export/import the content into Revit.  Or it's seems possible (also) to import the MP data point cloud into Revit directly without RECAP. 

QUESTION #1: Let's say a person scans an existing building in 3D with the MP camera, the rooms, to refurbish premises, you now, to redesign the premises in Revit.  Is there a way to convert the MP pointcloud into Revit 3D directly or you basically need to redraw everything in 3D over the pointcloud to make it usable?

For instance we have an existing building and the goal is to have it in 3D, BIM, scanned with MP in Pointcloud, is it the right way to do it?

QUESTION #2: Is it possible to get 2D floor plans out of the point cloud and how much processing is required in Revit to do that? Is there an option to do it automatically with minimal processing and time? of it's required to redraw the whole 3D over (in?) the point cloud in order to export a 2D floor plan in DWG? 

This is really preliminary survey, I never worked with pointclouds or Revit yet, it's going to happen somewhere in 2024. 

We have some time but I am curious. 

Thanks. 

[vitorbortoncello]

Question #1: Converting Matterport Point Cloud to Revit 3D**

Yes, it is possible to convert Matterport point clouds into a usable format in Revit. There are a few approaches to achieve this:

1. **Using Recap**: As you mentioned, you can use Autodesk ReCap to process the MP 3D data and texture and then import the result into Revit. ReCap can help clean up the point cloud data and make it more manageable for Revit.

2. **Direct Import**: Alternatively, some versions of Revit allow for direct import of point cloud data. You may need to check if your version supports this feature. Direct import can streamline the process by eliminating the need for an intermediate step with ReCap.

Regardless of the method chosen, you'll typically need to trace over the point cloud in Revit to create the actual 3D geometry. While the point cloud provides a reference for accurate positioning and dimensions, it doesn't automatically create the 3D model for you. Redrawing is often necessary to create the BIM model.

Question #2: Generating 2D Floor Plans from Point Cloud**

Yes, it is possible to generate 2D floor plans from a point cloud in Revit. However, this process also requires some manual work:

1. **Tracing Over Point Cloud**: Similar to creating a 3D model, you'll need to trace the walls, doors, windows, and other features over the point cloud to create the floor plan in Revit.

2. **Automated Tools**: Revit does offer some automated tools for creating 2D drawings from 3D models, such as the "Create Orthographic View" tool. While these tools can speed up the process, they may not always provide the desired level of accuracy, especially with complex geometries.

3. **Redrawing**: Depending on the complexity of the building and the desired level of detail in the floor plan, you may need to redraw certain elements to ensure accuracy.

In summary, while working with Matterport point clouds in Revit can streamline the process of creating 3D models and floor plans of existing buildings, it still requires a significant amount of manual work to create usable BIM models. However, the accuracy and detail provided by the point cloud data can greatly facilitate the modeling process compared to starting from scratch.

[Anonymous]

Hello thanks for the info.  

I should ask this question to matterport but I would like to ask a Revit user instead. When they scan a floor with matterport, do they need to spend time processing each scanned section manually? joining the rooms and corridors together and could this induce a lot of errors in wall thickness for instance, caused by poor fittings and other errors from the operator?

Or basically these cameras have some sort of GPS built in and they exactly know where they are at all time on a floor, so they can fit or join the scanned rooms together automatically ? And How good is the technology at doing it? 

I am pretty sure that a room can be super precise, but what about the whole picture of the floor? the camera has to be moved all day long, it can either build the floor automatically by joining the scanned rooms or it's done manually by the operator.  

If we spend a lot of time redrawing the whole point cloud of a floor on Revit, What I am interested to know if the floor is perfectly laid out or if there could be a lot of operator generated errors during the Matterport scanning process.

Before we spend a couple of grands on this. 

Thanks.