About the processed data set used in this implementation

erikwijmans / Pointnet2_PyTorch

PyTorch implementation of Pointnet2/Pointnet++

The Unlicense

1.53k stars 347 forks source link

About the processed data set used in this implementation #98

Closed CodeLHY closed 4 years ago

CodeLHY commented 4 years ago

hi @erikwijmans , I'm sorry to bother you again, I noticed that you use the processed data set(the first three dim is XYZ, and the rest are called "feature" in your code) instead of the origin ModelNet40. How can I know what the process of the original data set is, or in other words how to calculate the "feature"?

erikwijmans commented 4 years ago

The data comes from XYZ+normals in the official TF implementation here: https://github.com/charlesq34/pointnet2#shape-classification