Asymmetrical UpSample Block. Args: in_channels (int): Input channels of the block. out_channels (int): Output channels of the blo

Asymmetrical Residual Block. Args: in_channels (int): Input channels of the block. out_channels (int): Output channels of the blo

Asymmetrical DownSample Block. Args: in_channels (int): Input channels of the block. out_channels (int): Output channels of the blo

Dimension-Decomposition based Context Modeling. Args: in_channels (int): Input channels of the block. out_channels (int): Output

r"""Evaluate panoptic results for Semantickitti and NuScenes. Please refer to the `semantic kitti api <https://github.com/PRBonn/semantic-kitt

Mask sampling result.

Asymmetrical 3D convolution networks. Args: grid_size (int): Size of voxel grids. input_channels (int): Input channels of the blo

Points / Image pre-processor for point clouds / vision-only / multi- modality 3D detection tasks. It provides the data pre-processing as foll

LaserMix data augmentation. The lasermix transform steps are as follows: 1. Another random point cloud is picked by dataset. 2.

Load Annotations3D. Load instance mask and semantic mask of points and encapsulate the items into related fields. Required Keys: -

A pseudo sampler that does not do sampling actually.

P3Former.

P3Former head for 3D panoptic segmentation.

PolarMix data augmentation. The polarmix transform steps are as follows: 1. Another random point cloud is picked by dataset. 2.

r"""SemanticKitti Dataset. This class serves as the API for experiments on the SemanticKITTI Dataset Please refer to <http://www.semantic-kit