pointcloud with 7 features

[Sottix99]

hello, my pointcloud data has 7 features (x,y,z,nx,ny,nz, A) where the first 3 are classical coordinates, the second chunck from 3 represents the normals and the last element is the triangle area associated with the point. to make the model defined by Menghao work, what should i change? i simply camped imput_dims = 7 but the model doesn't work very well in perfromances

[qq456cvb]

As the original model works with the xyz coordinates, can you try feeding your data with only xyz coords. I would expect the triangle area be a disturbing factor since one can always simpify an object(thus increasing the area) but still keep the object roughly the same. Another thing is that you may want to check if your xyz coordinates spread uniformly across the surface in order to generalize.

[qq456cvb]

Setting input_dim=7 on its own won't do it. dataset.py hardcodes loading 3 or 6 channels (data[:, 0:3] / data[:, 0:6]), so the 7th column never reaches the model. You'd need to load the extra column in PartNormalDataset.__getitem__, set input_dim to match, and normalize the area value. That last part matters: area is on a very different scale from coords and normals, so if you leave it unnormalized it tends to dominate and hurt training. Once it's actually loaded and scaled, results should be better.