KPConv-PyTorch

KPConv-PyTorch copied to clipboard

Efficiency of a large-scale point cloud prediction

2

Hi, Thomas! Thank you for your very complete source codes. Although I am a layman about 3D semantic segmentation, I tried your source codes and successfully applied KPConv to different...

maosuli

Replacing NearestUpsample layer

1

Hello, in Section 3.4 of the paper, while describing the KPFCNN architecture, it is mentioned that "It is possible to replace the nearest upsampling operation by a KPConv, in the...

deeptibhegde

config for different datasets

3

Hello, thank you for your innovative work. I would like to ask how the config of different datasets is determined?

houyongkuo

Why the cuda memory always change？

1

Why the cuda memory always change？Do you have any way to pin the memory size？The server I use is public, and when others run the program, my program will run...

luzonghao1

Custom dataset is not small enough to be stored in-memory

6

Hello ! Thanks for your open source code. I'm trying to train KPConv on my own dataset, and I'm having memory problems. In your S3DIS.py dataloader, I noticed that you...

camillelhenry

ModuleNotFoundError: No module named 'cpp_wrappers.cpp_neighbors.radius_neighbors'

6

Hi, Thank you for sharing your code. I am trying to use this project but I am getting the above error. I am using windows 10 and I have compiled...

abbasihsn

AttributeError: module 'cpp_wrappers.cpp_subsampling.grid_subsampling' has no attribute 'subsample'

3

Hi Thomas, Thank you so much for your fantastic work and detailed documentation. I'm trying to get started with KPConv-PyTorch using the pretrained weights on S3DIS you kindly included. I...

cucalyptas

ValueError: Truth values are stored in a 0D array instead of 1D array

4

Traceback (most recent call last): File "E:/3DPC/KPConv-PyTorch-master/train_SemanticKitti.py", line 324, in trainer.train(net, training_loader, test_loader, config) File "E:\3DPC\KPConv-PyTorch-master\utils\trainer.py", line 274, in train self.validation(net, val_loader, config) File "E:\3DPC\KPConv-PyTorch-master\utils\trainer.py", line 292, in validation self.slam_segmentation_validation(net,...

DerekSunYH

Thank you for your work, may I ask, what is the main implementation principle of the method of batch and neighbor calibration? Can't read the code very well.

swzaaaaaaa

Hello, about the random kernel point selection

1

Hello Hugues, I would really appreciate it if you are able to help me with the following two questions. https://github.com/HuguesTHOMAS/KPConv-PyTorch/blob/73e444d486cd6cb56122c3dd410e51c734064cfe/kernels/kernel_points.py#L109 Is there a specific reason why you set a normal...

longxiangxu321