Skip to content
This repository was archived by the owner on Feb 3, 2024. It is now read-only.

id_choice in dataset.py is not defined #6

Closed
yuchenlichuck opened this issue May 27, 2021 · 6 comments
Closed

id_choice in dataset.py is not defined #6

yuchenlichuck opened this issue May 27, 2021 · 6 comments

Comments

@yuchenlichuck
Copy link

I found that if we class_choice and use segmentation there is an id_choice, which is not defined. I think It may be the indices. could you check with it?

@yuchenlichuck
Copy link
Author

BTW, Thanks for your excellent work, it largely improved my data loading efficiency. Thank you so much!

@yuchenlichuck
Copy link
Author

And Do you have the data that concentrate the XYZ axis with the normal vector (6-dim)?

@antao97
Copy link
Owner

antao97 commented May 28, 2021

Thanks for your feedback! I have fixed the bug and uploaded the code.

Note that id_choice is not indices. See my new code and antao97/dgcnn.pytorch#10 for details.

I don't have data with normal vectors. You can generate your own dataset with normal vectors.

@yuchenlichuck
Copy link
Author

Do you know how to get the normal vectors from the original dataset?

@antao97
Copy link
Owner

antao97 commented May 30, 2021

Since the point clouds are obtained on mesh, you can calculate the normal vectors on the mesh and then cast them into point clouds.

@yuchenlichuck
Copy link
Author

Sure, thanks a lot

@antao97 antao97 closed this as completed May 31, 2021
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants