open3d.ml.torch.datasets

Classes

Argoverse(dataset_path[, info_path, name, ...])	This class is used to create a dataset based on the Agroverse dataset, and used in object detection, visualizer, training, or testing.
Custom3D(dataset_path[, name, cache_dir, ...])	A template for customized dataset that you can use with a dataloader to feed data when training a model.
InferenceDummySplit(inference_data)
KITTI(dataset_path[, name, cache_dir, ...])	This class is used to create a dataset based on the KITTI dataset, and used in object detection, visualizer, training, or testing.
Lyft(dataset_path[, info_path, name, ...])	This class is used to create a dataset based on the Lyft dataset, and used in object detection, visualizer, training, or testing.
MatterportObjects(dataset_path[, name, ...])	This class is used to create a dataset based on the Matterport-Chair dataset and other related datasets.
NuScenes(dataset_path[, info_path, name, ...])	This class is used to create a dataset based on the NuScenes 3D dataset, and used in object detection, visualizer, training, or testing.
ParisLille3D(dataset_path[, name, ...])	This class is used to create a dataset based on the ParisLille3D dataset, and used in visualizer, training, or testing.
S3DIS(dataset_path[, name, task, cache_dir, ...])	This class is used to create a dataset based on the S3DIS (Stanford Large-Scale 3D Indoor Spaces) dataset, and used in visualizer, training, or testing.
Scannet(dataset_path[, name, cache_dir, ...])	Scannet 3D dataset for Object Detection, used in visualizer, training, or test.
SemSegRandomSampler(dataset)	Random sampler for semantic segmentation datasets.
SemSegSpatiallyRegularSampler(dataset)	Spatially regularSampler sampler for semantic segmentation datasets.
Semantic3D(dataset_path[, name, cache_dir, ...])	This class is used to create a dataset based on the Semantic3D dataset, and used in visualizer, training, or testing.
SemanticKITTI(dataset_path[, name, ...])	This class is used to create a dataset based on the SemanticKitti dataset, and used in visualizer, training, or testing.
ShapeNet(dataset_path[, name, ...])	This class is used to create a dataset based on the ShapeNet dataset, and used in object detection, visualizer, training, or testing.
SunRGBD(dataset_path[, name, cache_dir, ...])	SunRGBD 3D dataset for Object Detection, used in visualizer, training, or test.
TUMFacade(dataset_path[, info_path, name, ...])
Toronto3D(dataset_path[, name, cache_dir, ...])	Toronto3D dataset, used in visualizer, training, or test.
Waymo(dataset_path[, name, cache_dir, use_cache])	This class is used to create a dataset based on the Waymo 3D dataset, and used in object detection, visualizer, training, or testing.

Modules

augment	Dataset augmentation classes with functions such as random rotation, translation, and scaling.
samplers	Various algorithms for sampling points from input point clouds.
utils	Utilities for processing data, such as normalization and cropping.