sapcu icon indicating copy to clipboard operation
sapcu copied to clipboard

A question about "KITTI experiment"

Open QWTforGithub opened this issue 1 year ago • 6 comments

Thank you very much for your well organized code. May I ask how the KITTI experiment was conducted? Do you divide KITTI into multiple blocks, then perform upsampling separately?

QWTforGithub avatar Jul 28 '23 03:07 QWTforGithub

yes, we divide the point cloud into blocks

xnowbzhao avatar Jul 28 '23 06:07 xnowbzhao

yes, we divide the point cloud into blocks

Thank you. May I ask if point cloud sampling is not divided into object level and scene level? A scene point cloud (KITTI) can also be divided into multiple blocks and upsampled before merging, right? Compared to regular object level point clouds (ShapeNet), is point clouds like KITTI more challenging?

QWTforGithub avatar Jul 28 '23 07:07 QWTforGithub

The main challenge is the generation of seed points, the proposed method is hard to process large-scale point cloud directly, thus we choose to divide it into small blocks.

xnowbzhao avatar Jul 28 '23 07:07 xnowbzhao

The main challenge is the generation of seed points, the proposed method is hard to process large-scale point cloud directly, thus we choose to divide it into small blocks.

Thank you for your reply. The existing point cloud sampling methods seem to be experiments conducted at the object level. Is it meaningful to divide point cloud sampling into scene level and object level? Because a scene can also be divided into multiple point cloud blocks as input, and then merged.

QWTforGithub avatar Jul 28 '23 07:07 QWTforGithub

Sorry, i have not seen any discuss about the divide method.

xnowbzhao avatar Jul 28 '23 08:07 xnowbzhao

Sorry, i have not seen any discuss about the divide method.

Thank you.

QWTforGithub avatar Jul 28 '23 08:07 QWTforGithub