Hi Daniel,
I understand how c2c, c2m, and m3c2 work, I think.
I know there are pros and cons of each method, especially as they relate to the quality of the data sets being compared.
For me, I have a very nice, high quality, UAV SfM derived cloud of my target as a reference.
Then I have very poor ground based SfM derived clouds that used low resolution cameras.
I'm wanting to compare them to the beautiful UAV cloud of course.
1. I've been using the UAV cloud to mesh, using Delaunay Triangulation.
I understand Poisson is better, but it's meant for closed 3D objects, whereas I have a terrain.
Am I right to lean away from Poisson?
2. I haven't tried m3c2 yet, but am planning to. Any reason I shouldn't in my case? The reference cloud is beautiful, but the others are....subpar...
3. In reading on M3c2, I wonder....why don't we compare mesh to mesh? Is the averaging introducing too much error if noise in the cloud is high?
Am I right to think M3c2 is better than mesh to mesh anyway as it takes its averages over tinier areas in the clouds. Does that make sense?
Thanks!!!
Lindsay