Lidar_AI_Solution icon indicating copy to clipboard operation
Lidar_AI_Solution copied to clipboard

Published 20 hours ago • verified publisher icon NVIDIA-AI-IOT

About voxelization parameters

Open catziyan opened this issue 1 year ago • 7 comments

The voxelization parameters I used during training are min_range: [-27.0, 0, -5.0] max_range: [27.0, 54.0, +3.0] voxel_size: [0.075, 0.075, 0.2]

and I successfully converted the trained model to ONNX. However, when I modified the voxelization parameter in the code according to the parameters used during training, I got incorrect inference results. Surprisingly, not modifying the parameters yielded better results. Why is this?

catziyan avatar Sep 03 '23 06:09 catziyan

Maybe you should double-check your training script to see what has happened.

hopef avatar Sep 04 '23 01:09 hopef

I am also meet this problem. When I use default parameters, I can obtain the correct result, but when I expand the "max_x_range", I also got wrong results.
I check the input of scn_engine. Only sparse_shape=[41, 1440, 1440] can got the correct result.

qyang1996 avatar Sep 06 '23 07:09 qyang1996

I'm sorry. I'll check it soon.

Could you provide the wrong code and model?

hopef avatar Sep 06 '23 07:09 hopef

I am using the CenterPoint model trained on my own data. training config is: voxel_generator = dict( range=[-30.0, -75.0, -5.0, 120.0, 75.0, 3.0], voxel_size=[0.075, 0.075, 0.2] ) The sparse shape will be[40, 2000, 2000].

When I modify these parameters in the common.h file, I got the incorrect 3D box. I compare the python reuslt of generateVoxels with this project, it is similar. When I manually update the sparse shape to [41, 1440, 1440] befor scn_engine_->forward(), I can get correct 3D box.

qyang1996 avatar Sep 06 '23 08:09 qyang1996

Is there any update about this issue?

san9569 avatar May 17 '24 02:05 san9569

I am using the CenterPoint model trained on my own data. training config is: voxel_generator = dict( range=[-30.0, -75.0, -5.0, 120.0, 75.0, 3.0], voxel_size=[0.075, 0.075, 0.2] ) The sparse shape will be[40, 2000, 2000].

When I modify these parameters in the common.h file, I got the incorrect 3D box. I compare the python reuslt of generateVoxels with this project, it is similar. When I manually update the sparse shape to [41, 1440, 1440] befor scn_engine_->forward(), I can get correct 3D box.

Did u transfer your data to nuscenes ? How?

ruishanyin avatar Jul 02 '24 08:07 ruishanyin

@qyang1996

ruishanyin avatar Jul 02 '24 08:07 ruishanyin