Open hcscysh opened 2 months ago
Hi @hcscysh, thanks for your interest. The dataset is huge and didn't fit my Google Drive, which will come later. Your PLY files need to be converted to VG files (you can do so in Mapple - Point Cloud
- RANSAC primitive extraction
to extract planar primitives and save to VG) before taken to the data_loader
with specified vertex_group_path
: https://github.com/chenzhaiyu/polygnn/blob/67addd77a6be1d100448e3bd7523babfa063d0dd/dataset.py#L64
Note that many changes (including docs) will come once the paper completes the long review procedure.
Since the beginning of this project, I have started to pay attention. The latest release has not updated url_dataset, the default is null. I want to know how to test my own point cloud data. My point cloud data type is .ply.