Point cloud of simple models

[K40555]

Does anyone know where I can get point cloud models of for instance a cylinder or something equally easy. In which I can edit some of the coordinates.

[daniel]

Well, in CC you can use the 'Primitive Factory' (File menu) to create a cylinder, then play with its parameters (directly in the entity properties) and eventually convert it to a point cloud with 'Edit > Mesh > Sample points'.

[K40555]

Well, in CC you can use the 'Primitive Factory' (File menu) to create a cylinder, then play with its parameters (directly in the entity properties) and eventually convert it to a point cloud with 'Edit > Mesh > Sample points'.

Once I have this point cloud, is there any way I can get a file with it's coordinates?

[K40555]

Well, in CC you can use the 'Primitive Factory' (File menu) to create a cylinder, then play with its parameters (directly in the entity properties) and eventually convert it to a point cloud with 'Edit > Mesh > Sample points'.

I also can't seem to save my point cloud nor use most of the functions such as the statistics tool.

[daniel]

To export the cloud, be sure to select it in the DB tree (the entity with the cloud icon, and not its parent group). Then you'll be able to save it (with 'File > Save') as an ASCII/text file for instance (easily readable by a lot of programs).

And the statistics tool only works for points clouds with one or several scalar fields (statistics are computed on scalar fields).

[K40555]

To export the cloud, be sure to select it in the DB tree (the entity with the cloud icon, and not its parent group). Then you'll be able to save it (with 'File > Save') as an ASCII/text file for instance (easily readable by a lot of programs).

And the statistics tool only works for points clouds with one or several scalar fields (statistics are computed on scalar fields).

Thank you for your help, really appreciated. I want to create a couple of cylinders lets say 4. And I want them to have slightly different measures, assign them both random and systematic errors. I guess the easiest way would be to create them myself. This is like a "test" before I hop on the real project where I will be assigned 4 or 5 clouds of the same material where I am going to analyze their differences.

How do you suggest this should be done? My biggest question is that comparing them pointwise isn't the smartest thing to do, since generating a point cloud of the exact same model will be having different points anyways. How do I compare them on a larger scale but still able to get numbers out of it?

[daniel]

Not sure to understand: your real project will involve comparing point clouds or meshes? Because in the first case, the random pattern of sampled points is definitely something you'll get with real data. Otherwise if you are extracting "macro" measurements, the random sampling shouldn't have a noticeable effect as well (as it shouldn't modify the statistics very much, especially if you consider the average or standard deviation).

And last but not least, if you don't have points / vertices, you won't be able to compare things in CloudCompare ;)