Issues
- 1
- 2
- 0
- 0
How do you visualize the two datasets, shapenetpat and scanobjectnn, and if there is a way to visualize them, please send them to my email 1697909415@qq.comis:issue is:open
#34 opened by jc991206 - 0
How do I run this network with my own data?
#33 opened by mostafa501 - 0
- 3
TypeError: knnquery_cuda()
#25 opened by 0belle - 0
Methods about how to render point clouds
#31 opened by LAB123-tech - 0
Can you provide a SUN RGB-D load and train code
#30 opened by baigirl - 5
Segmentation Validate Stalls
#28 opened by amirhszd - 2
about flops and params
#27 opened by mmiku1 - 12
- 0
经过伞变换之后的值会很小,经过一些relu和卷积操作会出现很多nan值?
#26 opened by Lizhijian-123 - 5
Export RepSurf to ONNX fails
#23 opened by jelenab98 - 4
problem of multiprocessing
#17 opened by Liuyunming1999 - 1
Problem of the precalculated class weight
#18 opened by ProgramWho - 1
segmentation for ScanNet
#21 opened by fenfenglitech - 3
您好,想问问单卡3090的超参数您试过没没啊
#22 opened by SC-shendazt - 11
pointops module - cuda extensions on windows
#20 opened by anassmu - 0
Testing without cuda / converting to onnx runtime
#19 opened by anassmu - 0
The problem about centroid in Umbrella
#16 opened by Shadow581 - 0
- 4
- 1
When I just use UmbrellaSurfaceConstructor in front of other networks, the training of the whole network becomes very slow
#13 opened by ybc-ybc - 3
When will you publish codes about scannetv2?
#9 opened by YehLi - 1
- 2
- 6
- 5
- 2
- 5
can you add 6-fold train and test code?
#10 opened by mmiku1 - 3
Where can I find "lib/pointops"?
#6 opened by JiachengDeng - 1
Thanks to the author for the excellent work, but I wonder why group_centriod = torch.zeros_like (sorted_group_xyz), why is the value of group_centriod all 0?
#5 opened by dengyingxu - 1
- 1
Backbone for segmentation
#2 opened by tranganhthuan