-
install.md has some issue:
pip install mmcv-full==1.4.0
python setup.py install
You can try it
-
Thanks for your great job
When I train the model with tiny_fp16, I find the followings bug:
Traceback (most recent call last): …
-
I have clone mmdetection3d and BEVFormer repository, and follow the installtion step as [this](https://github.com/fundamentalvision/BEVFormer/blob/master/docs/install.md). My files tree is as follow…
-
Hi, @TempleX98 @Sense-X @CaraJ7
Thanks for your great work!
I'd like to know any plan to release code for HoP on BEVFormer?
-
Hello there,
When I was running BEVFormer model, I met this error:
![image](https://github.com/PaddlePaddle/Paddle3D/assets/49177682/11d89246-b9f7-40d3-8347-462236739000)
Does anyone know where's…
-
Thanks for yout great work!
I am doing evaluation on the temporal setttings, but obtain lower performance than that you have reported in github.
**My command:**
```bash
bash tools/dist_test_ma…
-
```py
# bev_mask = bev_mask.permute(2, 1, 3, 0, 4).squeeze(-1)
bev_mask = bev_mask.permute(2, 1, 3, 0, 4)[..., 0]
```
-
Hi. Thank you for the great work. As mentioned in https://github.com/zhiqi-li/BEVFormer/issues/34, is there any new information about the code of BEVFormer++. I'm looking forward to learning from your…
-
bevformer代码在去年五月左右开源。
作者为什么先采用IPM生成bev特征,然后使用类似于Deter-head检测地图元素,而不是使用bevformer直接检测出地图元素呢,请问是因为效果不好,还是没有尝试过呢?
-
https://github.com/zhiqi-li/BEVFormer/blob/df4023928133d3aa56c96c8185b66f15511bd86e/projects/mmdet3d_plugin/bevformer/modules/spatial_cross_attention.py#L137-L142
It seems to generate indexes for eac…