creating mesh from 3D-scanned objects
Posted: Fri Dec 06, 2019 7:03 am
Hi there,
I just downloaded cloudcompare yesterday from a recommendation to do my project. I don't really have a lot of idea how to use the software, but I am learning through YouTube and this forum.
I have a 3D scanner and I am trying to scan mangrove trees/forests and print a portion of it with a 3D scanner.
Based on what I've learned, I opened my point-based data, estimated the normal (though I have no clue what this means) using the PCL wrapper plugin (didn't change the search radius), and did Poisson reconstruction.
The result is something I didn't expect? I think because I scanned on the outdoors, there was a lot of noise in the data, but I don't know how to reduce/eliminate that (yet).
I have attached some pictures. I did the Poisson reconstruction with octree depth (don't what this means either) of 5 and 10.
Hopefully it's not too much trouble to help a newbie out. Any recommendations, advice or even links to tutorials regarding this topic would be very much appreciated.
Best,
Rinaldi
I just downloaded cloudcompare yesterday from a recommendation to do my project. I don't really have a lot of idea how to use the software, but I am learning through YouTube and this forum.
I have a 3D scanner and I am trying to scan mangrove trees/forests and print a portion of it with a 3D scanner.
Based on what I've learned, I opened my point-based data, estimated the normal (though I have no clue what this means) using the PCL wrapper plugin (didn't change the search radius), and did Poisson reconstruction.
The result is something I didn't expect? I think because I scanned on the outdoors, there was a lot of noise in the data, but I don't know how to reduce/eliminate that (yet).
I have attached some pictures. I did the Poisson reconstruction with octree depth (don't what this means either) of 5 and 10.
Hopefully it's not too much trouble to help a newbie out. Any recommendations, advice or even links to tutorials regarding this topic would be very much appreciated.
Best,
Rinaldi