C2C absolute distances

[lozanda]

Hi all,
I´m pretty new with CloudCompare. I made Cloud to Cloude compare with a nice result right now.
But I´m wondering why I can see only positive (absolute) values in C2C absolute distances. I´d like to see if points I´m comparing with reference point cloud are above or under the reference. I can see relative values in X,Y and Z but I´m just wondering if it´s possible to do the same with distances?
Thanks for any advice or just explanation....

[daniel]

It's because by default point clouds don't give the information about where is the inside and where is the outside (moreover with the closest point distance, we don't know how to flag the compared points that are on the surface but between two points in the reference cloud for instance).

To do this you'll need to use qM3C2. It's a much more advanced method that will let you compute signed distance between points clouds. The other solution is to mesh the reference cloud.

[lozanda]

OK, now I understadn - I´m gonna try to use qM3C2 first or mesh if it´s not gonna work fine for me :-)
Anyway thanks for advicce and explanation!
/Lenka

[Joey2780]

Hi Daniel,

I am new to CloudCompare myself, I am trying to obtain the Differences in X, Y and Z from the C2C test. I have looked through some other questions and this one seems most relevant to mine. I am trying to determine if there is a constant error in the X or Y or Z axis from point clouds.

Thank you!

[daniel]

Ok, good luck ;). Or did I miss your question?

[rozi]

Hi, I am also new to cloud compare. I am using version 2.6.2 and I was wondering what are my options for calculating real values for cloud to cloud (or cloud to mesh) distances in x,y,z directions. I am comparing two clouds and I need to know if the difference is positive or negative. When I do cloud to mesh comparison I get just one single value not split into x,y,z. Is the plugin my only option?

Thanks a lot! Btw, very nice software.

[daniel]

You can't split the distance in the case of Cloud-to-Mesh distances. But of course you could sample a lot of points on your mesh (so as to get a very good density) and then use the resulting cloud as reference for Cloud-to-Cloud distances.

[rozi]

Thank you Daniel, I will try that.