Can I scan the space more than 10m


#1

I want to scan the room and get their point cloud. Therefore, I am learning demo APP room capture.

However, I noticed that there is a limitation about room size. When I enlarge volume size in code. it showed

You need to specify volume bounds (kSTMapperVolumeBoundsKey) and volume resolution such that the resulting volume size in each dimension doesn’t exceed 10 meters.

Is there any way that I can scan the whole big room for one time? or I can scan many parts of the whole room and combine them?

Thanks


#2

Due to the RAM limitations on some supported iOS devices, the largest size that can be scanned is roughly 10m^2.

You can, however, combine the scans taken and we usually recommend using a 3D graphics program to help, like Blender, MeshLab, or SketchUp.

Theoretically, you could combine multiple scans together after they have been saved (see writeToFile:options:error: within STMesh class), though we don’t provide a way to do this in the SDK.


#3

Thanks for your reply.

Btw, since I want to scan larger room. Is it possible to scan every part of a room to produce mesh and combine them using via Transformation Matrix?

I looked up the code for Room Capture, there is a function called void MeshRenderer::uploadMesh(STMesh* mesh) and it may have multiple Meshes.

Can I use them to combine multiple meshes from different parts of a room? But I don’t know how to define their relative position.

Thanks


#4

We don’t provide the functionality for combining multiple meshes into one within the Structure SDK and MeshRenderer::uploadMesh(STMesh* mesh) is used to load a single mesh.

If this is some functionality you’d like to have, combining them based on some relative position will have to be designed by you.

I would specifically take a look at the STTracker class for information on how we store positional data.