You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Just curious what's the main difference between get_surface_sliding and get_surface_occupancy? I know that the latter one uses a level set value of 0.5 because it is extracting a mesh from an occupancy network. But can I also just use get_surface_sliding with a level set value of 0.5 for occupancy based network like unisurf?
The text was updated successfully, but these errors were encountered:
Hi, get_surface_sliding use multi-scale strategy to skip empty space since if sdf is very large, it means a large region centred at this point is empty. However, occupancy only have the inside outside information, the multi-scale strategy might not work well.
Just curious what's the main difference between get_surface_sliding and get_surface_occupancy? I know that the latter one uses a level set value of 0.5 because it is extracting a mesh from an occupancy network. But can I also just use get_surface_sliding with a level set value of 0.5 for occupancy based network like unisurf?
The text was updated successfully, but these errors were encountered: