Hi, Thanks for your good job!
I am trying to reproduce the result(mIoU=41.21) of BEVFormer w/OSP, but can not find the correct config. In your README-Results, OSP and BEVFormer w/OSP share the same config(projects/configs/osp.py), could you please provide the detailed config for BEVFormer w/OSP(e.g., the config to freeze trained BEVFormer)? Besides, I have some other questions:
there are 3 encoder layers, but each layer forward twice BEVFormerEncoder, why? Any explanation is appreciated.
i find a hard-code use_bev_volume=False, is this should be True for using OSP as a plugin to argument volume-based methods? and how to modify the code?
Hi, Thanks for your good job! I am trying to reproduce the result(mIoU=41.21) of BEVFormer w/OSP, but can not find the correct config. In your README-Results,
OSP
andBEVFormer w/OSP
share the same config(projects/configs/osp.py), could you please provide the detailed config forBEVFormer w/OSP
(e.g., the config to freeze trained BEVFormer)? Besides, I have some other questions:use_bev_volume=False
, is this should beTrue
for using OSP as a plugin to argument volume-based methods? and how to modify the code?Looking forward to your reply, Thanks!