Open
Description
The torchsparse papers seem focused on outdoor lidar datasets, which are very sparse. I don't see any benchmarks on slightly denser datasets like 3DMatch, where the point clouds are indoor spaces collected with a ToF camera or structured light.
Does torchsparse offer significant efficiency gains in these cases? Or is torchsparse focused on outdoor lidar?
Metadata
Metadata
Assignees
Labels
No labels