Boxes3D

Represents 3D axis-aligned or oriented bounding boxes defined by half-extents, centers, and optional Euler-angle rotations. This datatype is used for 3D regions of interest, object extents, and operations such as filtering by bounding box within the Telekinesis ecosystem.

Field

Required

Field	Type	Description
half_sizes	ArrayLike	Half-extents of each box; shape (3,) or (N, 3), dtype float32.
centers	ArrayLike	Centers of each box; shape (3,) or (N, 3), dtype float32.

Optional

Field	Type	Description
rotations_in_euler_angle	ArrayLike	Rotations in Euler angles (radians or degrees per implementation); shape (3,) or (N, 3), dtype float32. Default [0, 0, 0].

half_sizes and centers must have the same number of rows. If rotations_in_euler_angle has one row, it is broadcast to all boxes.

Methods

Method	Description
to_dict()	Returns a dictionary with keys: half_sizes, centers, and optionally rotations_in_euler_angle. Values are the stored arrays.

Example

From the datatypes examples:

from datatypes import datatypes
from loguru import logger

# ------------------------------------------------
# 1. Create Boxes3D instance
# ------------------------------------------------
boxes_3d = datatypes.Boxes3D(
    half_sizes=[[1, 2, 3], [4, 5, 6], [7, 8, 9]],
    centers=[[0.0, 0.0, 0.0], [1.0, 1.0, 1.0], [2.0, 2.0, 2.0]],
    rotations_in_euler_angle=[[0.0, 0.0, 0.0]],
)

# ------------------------------------------------
# 2. Access data via to_dict
# ------------------------------------------------
data = boxes_3d.to_dict()
logger.info("Boxes3D to_dict half_sizes={}", data["half_sizes"])
logger.info("Boxes3D to_dict centers={}", data["centers"])
logger.info("Boxes3D to_dict rotations_in_euler_angle={}", data["rotations_in_euler_angle"])