open3d.pipelines.registration.PoseGraphEdge

class open3d.pipelines.registration.PoseGraphEdge

Edge of PoseGraph.

__init__(*args, **kwargs)

Overloaded function.

  1. __init__(self: open3d.cpu.pybind.pipelines.registration.PoseGraphEdge) -> None

Default constructor

  1. __init__(self: open3d.cpu.pybind.pipelines.registration.PoseGraphEdge, arg0: open3d.cpu.pybind.pipelines.registration.PoseGraphEdge) -> None

Copy constructor

  1. __init__(self: open3d.cpu.pybind.pipelines.registration.PoseGraphEdge, source_node_id: int = -1, target_node_id: int = -1, transformation: numpy.ndarray[float64[4, 4]] = array([[1., 0., 0., 0.],

    [0., 1., 0., 0.], [0., 0., 1., 0.], [0., 0., 0., 1.]]), information: numpy.ndarray[float64[6, 6]] = array([[1., 0., 0., 0., 0., 0.], [0., 1., 0., 0., 0., 0.], [0., 0., 1., 0., 0., 0.], [0., 0., 0., 1., 0., 0.], [0., 0., 0., 0., 1., 0.], [0., 0., 0., 0., 0., 1.]]), uncertain: bool = False, confidence: float = 1.0) -> None

property confidence

Confidence value of the edge. if uncertain is true, it has confidence bounded in [0,1]. 1 means reliable, and 0 means unreliable edge. This correspondence to line process value in [Choi et al 2015] See core/registration/globaloptimization.h for more details.

Type

float from 0 to 1

property information

Information matrix.

Type

6 x 6 float64 numpy array

property source_node_id

Source PoseGraphNode id.

Type

int

property target_node_id

Target PoseGraphNode id.

Type

int

property transformation

Transformation matrix.

Type

4 x 4 float64 numpy array

property uncertain

Whether the edge is uncertain. Odometry edge has uncertain == false, loop closure edges has uncertain == true

Type

bool