hlesmqh/ws3d

Official version of 'Weakly Supervised 3D object detection from Lidar Point Cloud'(ECCV2020)

ws3d's Issues

trainning environment

What is the minimum training environment requirement( the required gpu)?

How to visualize your labelled BEV annotation ?

can you provide code to visualize your BEV click annotation?

where i can get label_w/label.txt?

Similar Paper

First I would like to say great work.

And I notice there is another paper called

Weakly Supervised 3D Object Detection from Point Clouds
https://arxiv.org/pdf/2007.13970.pdf

which is very similar with yours. So what's the main difference?
I'll try to figure out the basic ideas of these two papers.

ModuleNotFoundError: No module named 'pointnet2_msg'

Hi, thank you for your interesting work!

I am running your code and encounter the error ModuleNotFoundError: No module named 'pointnet2_msg'. So I wonder if you know how to solve that problem?

Thank you so much for your time.

Projecting from BEV to Point Cloud

How to generate the Pseudo front view by projecting from BEV into Pointcloud?
Can you provide code for that projection only?

No such file or directory: '/home/rtx-3090/dataset/Kitti/object/ImageSets/small_val.txt'

Hi! I cant find the small_val.txt in this project. How can I get it?

where is the label_w/label.txt

I cant find the label_w/label.txt. Can you provide more information about it?

how to eval the model?

thanks for your work, but I hard to know how to eval your trained model ? When I run the code "eval_auto" and "eval_active", it stuck. can you give a detailed instruction for the model eval?

How can I get the BEV center-click annotation, I can not open the link.