Pointnet_Pointnet2_pytorch
Pointnet_Pointnet2_pytorch copied to clipboard
yanx27
PointNet and PointNet++ implemented by pytorch (pure python) and on ModelNet, ShapeNet and S3DIS.
Hi, thank you for the wonderful work on pointnet. I try to use the shapenet dataset you provided in your project(https://shapenet.cs.stanford.edu/media/shapenetcore_partanno_segmentation_benchmark_v0_normal.zip). But I cannot find any usage guide on the...
您好，通常在预测点云语义分割的时候，只有点本身的数据，没有点的真实标签。但是在语义分割代码中使用了点的真实标签用来计算标签的权重，在投票的时候使用到了标签的权重，这样的设计是否合理呢？
Excuse me When I try to run _train_partseg.py_ with default parameters, I am stuck in `classifier = MODEL.get_model(num_part, normal_channel=args.normal).cuda()`. What could possibly be the reason? CUDA version is wrong?
I trained and tested myself data, I found a problem that preprocessing the data with "random. Choice" has a great impact on the results . I know that 'farthest_point_sample' function...
Thank you for sharing your code. I found that in the point cloud segmentation task, the training (train_semseg.py) is in the S3DIS dataset, but the test (test_semseg.py) is in the...
In the case of semantic segmentation: After running the command **(_python collect_indoor3d_data.py_)**, no files are there in the designated folder. The processed data should come but nothing is coming. Kindly...
in the PartNormalDataset loader it is different with the modelnetloader which uses the random sample instead of farthest_point_sample, I was very confused, can anyone help me to explain it? the...
Hi there, for the purpose of profiling the network, I modify the training bit so that it is wrapped in an "with ... as profiler" head following the Pytorch tutorial....
As mentioned in paper, we use skip connection in feature propagation layer, but I'm confused by this line in pointnet2_part_seg_msg.py : self.fp1 = PointNetFeaturePropagation(in_channel=150+additional_channel, mlp=[128, 128]) How do you get...
