activevisionlab / ninormal Goto Github PK

Dear author,

I wish to test point clouds with less than 100k points. Is there a way to pre-process point clouds with < 100k points to h5py? How should I change the script?

Thanks in advance!

I'd like to use a pretrained model on a specific point cloud (not in the test data) for estimating the normals. I'm not sure how to do this as I'm new to this and the README only explains how to achieve the results presented in the paper with the test data.

So far, I've tried to substitute the test data with my own so I created an h5 file with the preprocessing script. Running the test.py script on that data with one of the pretrained models returns a KeyError: "Unable to open object (object '0000000044' doesn't exist)". I really don't know what's wrong here or if I'm on the right track. On the internet, I've found that maybe something is wrong with the h5 file. Do you have any tips for me? Maybe there's a way to use the pretrained model without a h5 file since I only want to process one point cloud.

When opening the sharepoint links there is an error message.