Skip to content

pose

pose #

The 6-DoF camera-pose JSON scalar.

Pose #

Bases: JSONScalar

A 6-DoF camera pose: world position, orientation, and pinhole intrinsics.

Where an Angle carries orientation alone, a Pose adds the camera's world-space position (position_x/position_y/position_z, in metres) — what a world-tracked session recovers and a 3D reconstruction needs. Together the position, the orientation quaternion, and the intrinsics (focal_x/focal_y/center_x/center_y at reference_width/reference_height) place every pixel as a metric ray in world space. Intrinsics are absent when the capturing device delivered no calibration.

Attributes:

Name Type Description
position_x float

World-space position x in metres.
position_y float

World-space position y in metres.
position_z float

World-space position z in metres.
quaternion_w float

Quaternion scalar part.
quaternion_x float

Quaternion x component.
quaternion_y float

Quaternion y component.
quaternion_z float

Quaternion z component.
focal_x float | None

Horizontal focal length in pixels, at reference_width/reference_height.
focal_y float | None

Vertical focal length in pixels.
center_x float | None

Principal point x in pixels.
center_y float | None

Principal point y in pixels.
reference_width float | None

Pixel width the intrinsics are expressed in.
reference_height float | None

Pixel height the intrinsics are expressed in.
tracking_state TrackingState | None

The tracker's confidence at capture, if reported.
time Annotated[datetime | None, Epoch]

The frame's capture time (epoch seconds on the wire), if reported.
Source code in capturegraph-lib/capturegraph/types/scalars/position/pose.py 
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
class Pose(JSONScalar, name="edu.cornell.pose"):
    """A 6-DoF camera pose: world position, orientation, and pinhole intrinsics.

    Where an ``Angle`` carries orientation alone, a ``Pose`` adds the camera's
    world-space position (``position_x``/``position_y``/``position_z``, in metres) —
    what a world-tracked session recovers and a 3D reconstruction needs. Together the
    position, the orientation quaternion, and the intrinsics
    (``focal_x``/``focal_y``/``center_x``/``center_y`` at
    ``reference_width``/``reference_height``) place every pixel as a metric ray in
    world space. Intrinsics are absent when the capturing device delivered no
    calibration.

    Attributes:
        position_x: World-space position x in metres.
        position_y: World-space position y in metres.
        position_z: World-space position z in metres.
        quaternion_w: Quaternion scalar part.
        quaternion_x: Quaternion x component.
        quaternion_y: Quaternion y component.
        quaternion_z: Quaternion z component.
        focal_x: Horizontal focal length in pixels, at
            ``reference_width``/``reference_height``.
        focal_y: Vertical focal length in pixels.
        center_x: Principal point x in pixels.
        center_y: Principal point y in pixels.
        reference_width: Pixel width the intrinsics are expressed in.
        reference_height: Pixel height the intrinsics are expressed in.
        tracking_state: The tracker's confidence at capture, if reported.
        time: The frame's capture time (epoch seconds on the wire), if reported.
    """

    position_x: float
    position_y: float
    position_z: float
    quaternion_w: float
    quaternion_x: float
    quaternion_y: float
    quaternion_z: float
    focal_x: float | None = None
    focal_y: float | None = None
    center_x: float | None = None
    center_y: float | None = None
    reference_width: float | None = None
    reference_height: float | None = None
    tracking_state: TrackingState | None = None
    time: Annotated[datetime | None, Epoch] = None