Shaoyuan Xie

@czy341181 Do you train BEVFormer-S from scratch and get the results of the paper?

I've corrected it. You only need to change the following argument to switch between temporal and non-temporal mode. ``` video_test_mode=False ``` Thanks for pointing out this issue.

Yes, that's correct. In the official BEVFormer paper, the authors trained two versions: BEVFormer and BEVFormer-S. The BEVFormer-S is trained without temporal information, i.e., ```video_test_mode=False``` for both training and testing....

Thanks for your attention. we are sorry currently we don't have plans to update this. If we plan to update it in the future, we will let you know.

We will follow the same evaluation protocol of SurroundOcc for the RoboDrive challenge. Please refer to this [link](https://github.com/weiyithu/SurroundOcc/blob/main/docs/data.md) for more details. We will use the generated occupancy labels with resolution...

Hi, thanks for your interest, The nuScenes-C dataset is generated using the validation set of the nuScenes dataset, which only contains 6019 images for each camera. You can refer to...

Sorry for the confusion, you can refer to corruptions/project/config/nuscenes_c.py Please also notice that this script is used to generate corruption for nuScenes dataset. For your need to generate corruption on...

Hi, the mmcv package is as follow: ```sh mmcv-full 1.4.0 pypi_0 pypi mmdet 2.14.0 pypi_0 pypi mmdet3d 0.17.1 pypi_0 pypi mmsegmentation 0.14.1 pypi_0 pypi ``` Hope it can help. It...

Hi @yangxue-1, Thanks for reaching out. It seems this error is due to incomplete nuScenes dataset. I double check the error on my side and find out the file you...

Hi @jingzhengli, You can try to install the pre-compiled version of `mmcv` by running: ``` pip install mmcv-full==1.4.0 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.9.0/index.html ``` Please replace the torch and Cuda versions to your...