-
Hi, thank you for your implementation of CenterPoint. However, I have a few questions regarding the [results](https://github.com/open-mmlab/mmdetection3d/blob/master/configs/centerpoint/README.md#cent…
-
I use one 32-lidar and two 16-lidar fused point cloud. and I use velodyne64 configuration. But It can not detect enough obstacles. Not as many as apollo5.5. Does the point pillar only support velodyn…
-
Thank you for your codebase, I met a bug when trying to train PointPillars on Waymo.
**Checklist**
1. I have searched related issues but cannot get the expected help.
2. The bug has not been fixe…
-
PointPillars export to onnx ？
-
Is tensorboard is used to visualize since all are log files.
-
Hi. This is a follow-up to https://github.com/open-mmlab/mmdetection3d/issues/29 . I have a single GPU and I would like to train PointsPillars on NuScenes using (for example)
python tools/train.py --…
-
Hello,
Can this code be used to train a model for Autoware's Point Pillars package ?
Thank you.
-
Hi, thanks for great work.
I used existing configs to reproduce some results in NuScenes and there was no problem.
I saw that you have provided some free anchor configs as well. I wanted to rep…
-
Training _PointPillars_ and _SECOND_ works fine,
but when I try to train _PartA^2Net_, I get this error message after addressing issue [#70](https://github.com/sshaoshuai/PCDet/issues/70):
```
Tra…
-
anyone know how to load a trained model?I am a baginner , i am confused with loading trained model,then running the model on myself data for days? It is hard for me to do this.Can you tell in details?