It is definitely something with the coordinate systems and I do suspect it probably falls more to the side of CloudCompare than Reality Capture but I thought I would try here.
When I export from RC:
- Coordinate System: Project output (EPSG 6491; MA State Plane)
- Transformation Preset: Default
When importing the mesh model into RC:
- Coordinate System: Project output (EPSG 6491; MA State Plane)
- Import Coordinate System: EPSG 6491; MA State Plane
- Transformation Preset: Default
So I’m maintaining the coordinate system between exports and imports.
I’m overwriting the .obj file that I first exported from RC since I was hoping that referencing the RCINFO file would help place it in the right spot. I did this after my attempts at saving a new .obj file from CC had the same results.
There are two attributes of CC that I think may be messing this up:
- CloudCompare is “unit-less”, so when you open a file that is in meters, 1m will read “1.0” and if you open a file that has used ft it will also read “1.0” when you measure 1’ ft, and
- When importing CC will transform the origin of your dataset so it is closer to the origin. Now, CC does promise to undo that transformation when you export, but since I’m converting the point cloud to a mesh, I’m wondering if it is forgetting to do this…
I guess just isolating the diagnosis and determining which system is causing this is what I’m looking for. Since the coordinate systems I’m using are consistent in RC, I think I’m going to have to explore this on the CloudCompare side of the coin.