hold on

FCOS3D Inference Error: AssertionError: loss log variables are different across GPUs!

2

When after inferencing and start evaling, I came cross this bug: Below are logs: 86%|########5 | 5173/6019 [00:09

train the model with custom dataset(different size of image)

3

I'd like to train m3d-model using my custom dataset.I rewrite the imdb_util.py and it can generate imdb rightly.However,my image(W=3384,H=2710) in the custom dataset is larger than images in KITTI. At...

When I read the paper, I have a confusion that what is the time in below chart represents?  Does the 161 ms here including the preprocess of image and...

About split dataset:"test"

1

The paper told that there are three datasets in the evaluation: kitti_split1(val1),kitti_split2(val2) and test . My question is that test dataset seems to use the kitti/training to train and use...

Why not add steps for monocular 3D detection in readme?

1

Hi, I notice that the code can be transformed to monocular easily as the paper said.Why not adding the approach for mono in readme?

AttributeError: module 'g2o' has no attribute 'SparseOptimizer'

2

Installing with sudo (I was okay with that since I was using it in a container) seemed to work. Naturally that would be in the system python. I don't quite...

RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation

1

Hi ,I am interested in your work. When I run `python tools/plain_train_net.py --config-file "configs/smoke_gn_vector.yaml"` there is a bug: ``` Traceback (most recent call last): File "tools/plain_train_net.py", line 100, in args=(args,),...

Could you please provide the code of RTM3D?

1

Thanks for your great work!!!. After reading two papers(RTM3D, KM3D), it seems that you only make the code of KM3D open-source, isn't it ? I wonder to compare the performance...

Could you please provide your code for monocular 3D object detection?

2

Or could your provide me some tips?

Thanks for your great work and it impresses me ! I wanna to have a try in my research. Specificly, I wanna to use ViT-small to replace the ImageNet pretrained...