search

laughtervv / SGPN

SGPN:Similarity Group Proposal Network for 3D Point Cloud Instance Segmentation, CVPR, 2018
MIT License
268 stars 62 forks source link

Gap test file #29

Open valvador92 opened 5 years ago

valvador92 commented 5 years ago

Hi! Does anyone know what is "gap = 5e-3 " in the test file. Why this value? How to calculate or find this value. I am trying to run SGPN with my own dataset and this value has an influence on the output.

Thank you in advance

lelouedec commented 4 years ago

I think this is so that the density of points matches the size of the blocks of 111m. So dividing the coordinates reduce or augment the density to match this block size. I am implementing SGPN for my dataset as well, how did you manage to code GroupMerging with label_bin for your own dataset ?

jianuo1128 commented 1 month ago

I think this is so that the density of points matches the size of the blocks of 1_1_1m. So dividing the coordinates reduce or augment the density to match this block size. I am implementing SGPN for my dataset as well, how did you manage to code GroupMerging with label_bin for your own dataset ?

There is a GroupMerging function in the test_utils.py that can be used