zehuichen123/bevdistill

[ICLR 2023] BEVDistill: Cross-Modal BEV Distillation for Multi-View 3D Object Detection

bevdistill's Issues

Bug Question

Hello! Could you please tell me the version of mmdet3d you used?
when I run, it shows like:
from numba.np.ufunc import _internal
SystemError: initialization of _internal failed without raising an exception

source code question

Hello!
Will the source code be released? If so, is there a specific time

How to enable CBGS?

Hi, I noticed there are some results obtained by using CBGS in your paper. Could you please tell me how to enable CBGS in your code?

About the pretrained model

Hi, thanks a lot for sharing your code. I find that the student model is trained from a pre-trained model "data/pretrain_models/fcos3d_r50_new.pth". Can you provide it, or tell us how to get it?

Thank you very much.

About BEVFormer-T, which paper is this model originally from?

I assumed BEVFormer-T corresponds to the configs in configs/video_bevformer/*.
What is the difference with the original BEVFormer?
Could you provide distillation training configs between original BEVFormer and teacher model?
Thank you.

Could you please provide the additional metrics in Table 1?

Thanks for your great work. I am also exploring this topic and want to cite your work. I need the additional metrics in Table 1, including mATE, mASE, mAOE, mAVE and mAAE of your method and the reproduced ones. Could you please provide them?