search

paulguerrero / pcpnet

Pytorch implementation of PCPNet
Other
169 stars 41 forks source link

why should noise-free and noise point normal vector be different in PCPNet? #17

Open RyuJaeSeok opened 4 months ago

RyuJaeSeok commented 4 months ago

To author

I think it's kind of idiot question to you about PCPNet Datasets. I encountered about difference between noise-free normal and noise normal. but some kinds of objects are same.... I thought that kinds of difference will be represent to preserve geometry structure. but i can't understand that should be same from noise case or noise-free case. so can you tell me why should noise pointcloud need different ground truth type normal vector.

Thank you

paulguerrero commented 4 months ago

Hi @RyuJaeSeok ,

I am not sure I fully understand your question, are you asking why the GT normals for the non-noisy point clouds and the noisy point clouds are different? The two should have the same normals for points that were originally sampled from the same positions on the ground truth surface.

I am not 100% sure exactly how we created the noisy point clouds anymore, but we probably re-sampled the surfaces randomly for the noisy version, so points with at same index in the list of points will probably originate from a different location on the ground truth surface, so that would be reason for them having different ground truth normals.