sangjin
sangjin
Hi @JingweiZhang12, Do you mean that the current version of BEVFusion doesn't work now? The reason I ask is that ,when running the demo, the voxelization doesn't returns anything (i.e....
Hi, I can't remember how I solve 'this' issue. But, I've installed spconv just using `pip install spconv-cuxxx`
I just used `pip install spconv-cuxxx`
To comment `from .feature_decorator import feature_decorator` in `mmdet3d/ops/feature_decorator/init.py` could be temporal solutions.
> When I ues bevfusion in our own dataset(only one camera too), I get a drop in the fusion model's performance too， have you found the reason?? No, I'm using...
> Hello, I am also trying to learn BEVFusion on KITTI dataset but I keep getting errors, can you provide me some help, I will be very thankful! Hi, to...
> > > When I ues bevfusion in our own dataset(only one camera too), I get a drop in the fusion model's performance too， have you found the reason?? >...
> Nice! And I have another question, the image size of kitti datasets(1392 x 512) is different from nuscenes(1280 x 720)，how to set the resize parameter in data augmentation??? First,...
> Thanks so much. When I train, the following error occurs: File "mmdetection3d/projects/BEVFusion/bevfusion/depth_lss.py", line 299, in forward cur_coords = cur_img_aug_matrix[:, :3, :3].matmul(cur_coords) IndexError: too many indices for tensor of dimension...
@Lcl159, oops, sorry I was confused about it. I do not modify the img_aug_matrix. If the input images whose shape is (B, N, C, H, W) are processed, the augmentation...