Struct PointCloud

pub struct PointCloud { /* private fields */ }

An ordered set of 3D points with optional per-point channels.

Order is meaningful: it is the caller’s, and every channel is indexed by it. Operations that reorder points must say so, because a consumer holding a parallel array outside the cloud would otherwise be silently desynchronised.

Implementations

impl PointCloud

pub fn new(points: Vec<Point3>) -> Result<Self, PointCloudError>

A cloud of positions, with no attributes.

Errors

Refuses a non-finite coordinate, naming the offending index.

pub fn empty() -> Self

An empty cloud.

Legitimate rather than exceptional: a query that filters everything out should return an empty cloud, not an error. Consumers that need points must check is_empty themselves.

pub fn with_normals(self, normals: Vec<Vec3>) -> Result<Self, PointCloudError>

Attach per-point normals.

Errors

Refuses a length mismatch, and a normal that is non-finite or too short to carry a direction.

pub fn with_colours(self, colours: Vec<Colour>) -> Result<Self, PointCloudError>

Attach per-point colours.

Errors

Refuses a length mismatch. Colour channels cannot be individually invalid, since every 8-bit value is meaningful.

pub fn with_intensities( self, intensities: Vec<Scalar>, ) -> Result<Self, PointCloudError>

Attach per-point intensities.

Errors

Refuses a length mismatch and a non-finite intensity. The range is not constrained: sensors report intensity on their own scale, and normalising it here would discard information the caller may need.

pub fn points(&self) -> &[Point3]

The points, in the caller’s order.

pub fn normals(&self) -> Option<&[Vec3]>

Per-point normals, if the cloud carries them.

pub fn colours(&self) -> Option<&[Colour]>

Per-point colours, if the cloud carries them.

pub fn intensities(&self) -> Option<&[Scalar]>

Per-point intensities, if the cloud carries them.

pub fn len(&self) -> usize

Number of points.

pub fn is_empty(&self) -> bool

Whether the cloud has no points.

pub fn has_normals(&self) -> bool

Whether every point carries a normal.

Reconstruction algorithms differ sharply on this: Poisson needs oriented normals, ball pivoting does not. Asking is cheaper than discovering it mid-operation.

pub fn bounds(&self) -> Option<(Point3, Point3)>

Axis-aligned bounds as (min, max), or None when empty.

Provided here because it needs no algorithm and every consumer wants it; anything beyond this belongs in a query crate.

Trait Implementations

impl Clone for PointCloud

fn clone(&self) -> PointCloud

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for PointCloud

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for PointCloud

fn default() -> PointCloud

Returns the “default value” for a type.

impl PartialEq for PointCloud

fn eq(&self, other: &PointCloud) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for PointCloud