Struct KdTree 

pub struct KdTree { /* private fields */ }

A kd-tree built over a cloud.

The tree is built in absolute coordinates, the same ones PointCloud::point returns. That makes distances bit-for-bit comparable with the brute-force reference; otherwise the agreement test would need a tolerance and would stop catching indexing mistakes.

Implementations

impl KdTree

pub fn build(cloud: &PointCloud) -> Result<Self, SpatialError>

Builds a tree over a cloud.

pub fn build_observed<F>( cloud: &PointCloud, progress: F, ) -> Result<Self, SpatialError>

where F: FnMut(usize, usize),

The same, reporting progress after each internal phase.

progress receives (completed, total) in phases: the coordinates are extracted, then the tree is built. Two steps is all the resolution there is — kiddo builds the tree in one opaque call, and on a large cloud that call is the larger half of the wait. Pretending otherwise would mean a progress bar that lies.

pub fn len(&self) -> usize

Number of indexed points.

pub fn is_empty(&self) -> bool

Whether the tree is empty.

Trait Implementations

impl NeighborSearch for KdTree

fn knn_into(&self, query: &Vector3<f64>, k: usize, out: &mut Vec<Neighbor>)

The k nearest points to query, written into out, whose previous contents are discarded.

fn radius_into( &self, query: &Vector3<f64>, radius: f64, out: &mut Vec<Neighbor>, )

Every point inside the ball of radius radius, written into out.

fn knn( &self, query: &Matrix<f64, Const<3>, Const<1>, ArrayStorage<f64, 3, 1>>, k: usize, ) -> Vec<Neighbor>

Allocating convenience wrapper over knn_into.

fn radius( &self, query: &Matrix<f64, Const<3>, Const<1>, ArrayStorage<f64, 3, 1>>, radius: f64, ) -> Vec<Neighbor>

Allocating convenience wrapper over radius_into.