Logic behind using sample_groups=[ dict(car=2), dict(truck=3), dict(construction_vehicle=7), dict(bus=4), dict(trailer=6), dict(barrier=2), dict(motorcycle=6), dict(bicycle=6), dict(pedestrian=2), dict(traffic_cone=2), ]

V2AI / Det3D

World's first general purpose 3D object detection codebse.

https://arxiv.org/abs/1908.09492

Apache License 2.0

1.5k stars 298 forks source link

Logic behind using sample_groups=[ dict(car=2), dict(truck=3), dict(construction_vehicle=7), dict(bus=4), dict(trailer=6), dict(barrier=2), dict(motorcycle=6), dict(bicycle=6), dict(pedestrian=2), dict(traffic_cone=2), ] #134

Closed DeepakVellampalli closed 3 years ago

DeepakVellampalli commented 3 years ago

How to use Det3D

Questions like:

What is the logic in using values for car=, truck=3 etc for the code line

Det3D/examples/cbgs/configs/nusc_all_vfev3_spmiddleresnetfhd_rpn2_mghead_syncbn.py

DeepakVellampalli commented 3 years ago

@poodarchu , could you please help on this logic. Do the needful

tojimahammatov commented 3 years ago

hi @DeepakVellampalli , these numbers are magnitudes for each category. The magnitude means number of instances placed into a point cloud sample. See the paper of cbgs (table 5) for more information. This is for gt augmentation.