Number of the point cloud with color texture

Questions & Answers
1

Hi Team,

Can you give me some advice on this question?

General Information

  • Product: C57

Problem / Question

  • When the number of columns in the PLY file output after ComputePointMap is X, or the number of columns in an array with NaN points removed, when RenderPointMap is performed the number of arrays is much less than X. Is this because of the specification?

  • Does this mean 1) the point cloud is thinned out based on the concept of voxels, or 2) points that are not visible due to viewpoints such as ViewPose are thinned out?

  • If it’s reason 2), I think color information will not be added unless you save the PLY after performing RenderPointMap. In order to add color information, points behind the point closest to the camera in XY positions in the generated point cloud cannot be rendered, so will points with the same or very close XY positions, but different only in Z depth, be deleted, leaving only the point closest to the camera?

Warm regards, Kenzo

2

Hi,

RenderPointMap intentionally changes the resolution by resampling the point cloud from the chosen perspective. How exactly the resampling works depends on the chosen rendering method and is described in the manual.

The resulting resolution is either the specified Size for telecentric projection or the resolution of the target camera when specifying Camera.

For the C-series you probably use RenderPointMap to transform the point cloud to the perspective of the color camera, so the point cloud will have the same resolution as the color camera (~2MP) afterwards. This is useful to align the 3D data with features from the color image.

3

Hi Daniel,

Thank you for your prompt reply.

Warm regards, Kenzo