cloudViewer::t::geometry::kernel::pcapartition Namespace Reference

Functions
std::tuple< int, core::Tensor >	PCAPartition (core::Tensor &points, int max_points)

Function Documentation

PCAPartition()

std::tuple< int, core::Tensor > cloudViewer::t::geometry::kernel::pcapartition::PCAPartition	(	core::Tensor &	points,
		int	max_points
	)

Partition the point cloud by recursively doing PCA.

Parameters

points	Points tensor with shape (N,3).
max_points	The maximum allowed number of points in a partition.

Returns

The number of partitions and an int32 tensor with the partition id for each point. The output tensor uses always the CPU device.

Definition at line 83 of file PCAPartition.cpp.

References AssertTensorDtypes, AssertTensorShape, cloudViewer::core::Tensor::Empty(), cloudViewer::core::Float32, cloudViewer::core::Float64, cloudViewer::core::Tensor::GetDataPtr(), cloudViewer::core::Int32, LogError, cloudViewer::utility::nullopt, and points.

Referenced by cloudViewer::t::geometry::TriangleMesh::PCAPartition(), and cloudViewer::t::geometry::PointCloud::PCAPartition().