Struct MinSquaresTriangulator

pub struct MinSquaresTriangulator { /* private fields */ }

This solves triangulation problems by simply minimizing the squared reprojection error of all observances.

This is a quick triangulator to execute and is fairly robust.

use cv_core::nalgebra::{Vector3, Point3, Rotation3, Unit};
use cv_core::{TriangulatorRelative, CameraToCamera, CameraPoint, Pose, Projective};
use cv_geom::MinSquaresTriangulator;

let point = CameraPoint::from_point(Point3::new(0.3, 0.1, 2.0));
let pose = CameraToCamera::from_parts(Vector3::new(0.1, 0.1, 0.1), Rotation3::new(Vector3::new(0.1, 0.1, 0.1)));
let bearing_a = point.bearing();
let bearing_b = pose.transform(point).bearing();
let triangulated = MinSquaresTriangulator::new().triangulate_relative(pose, bearing_a, bearing_b).unwrap();
let distance = (point.point().unwrap().coords - triangulated.point().unwrap().coords).norm();
assert!(distance < 1e-6);

Implementations

impl MinSquaresTriangulator

pub fn new() -> Self

Creates a MinSquaresTriangulator with default values.

Same as calling Default::default.

pub fn epsilon(self, epsilon: f64) -> Self

Set the epsilon used in the symmetric eigen solver.

Default is 1e-9.

pub fn max_iterations(self, max_iterations: usize) -> Self

Set the maximum number of iterations for the symmetric eigen solver.

Default is 100.

Trait Implementations

impl Clone for MinSquaresTriangulator

fn clone(&self) -> MinSquaresTriangulator

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MinSquaresTriangulator

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

Formats the value using the given formatter.

impl Default for MinSquaresTriangulator

fn default() -> Self

Returns the “default value” for a type.

impl TriangulatorObservances for MinSquaresTriangulator

fn triangulate_observances<B: Bearing>( &self, pairs: impl IntoIterator<Item = (WorldToCamera, B)>, ) -> Option<WorldPoint>

impl Copy for MinSquaresTriangulator