robokudo.descriptors.camera_configs.config_cv_camera_without_depth

This is the camera config for a openCV camera without depth information.

Classes

OpenCVCameraConfig

Configuration class for OpenCV-based cameras without depth information.

Module Contents

class robokudo.descriptors.camera_configs.config_cv_camera_without_depth.OpenCVCameraConfig

Bases: robokudo.descriptors.camera_configs.base_camera_config.BaseCameraConfig

Configuration class for OpenCV-based cameras without depth information.

This class defines the configuration parameters for cameras that can be accessed through OpenCV’s video capture interface. It supports various input sources including physical cameras, video files, image sequences, and network streams.

Note

When using image sequences as input, the first image must have a number between 0 and 4, and there cannot be any gaps in the sequence numbering.

registry_name: typing_extensions.ClassVar[str] = 'opencv'

Name of the camera configuration in the registry. Must be unique.

interface_type: str = 'OpenCV'

Type of camera interface

device: typing_extensions.Union[int, str] = 0

Input source identifier (camera index, file path, or URL).

device (integer for I/O device or path to image/video file) and flag integer: id of the video capturing device to open

Use 0 to open default camera using default backend

string: path to a video file

path to image sequence (e.g. ‘my/path/img_%02d.jpg’).

Note: First image needs number between 0 and 4 and the following numbers cannot contain any gaps.

URL of video/camera stream or image

api_preference: int

Preferred OpenCV capture API backend

device_driver_flag: int = 0

Flag argument to use when retrieving/reading the frames

normalize_rgb: bool = True

Normalize/stabilise rgb image contrast/brightness

loop_mode: int = -1

Loop after iterating over all frames of the video file > 0: number of repetitions before to stop = 0: never loop < 0: infinite loop

… note::

Only works with a file as source

depth: typing_extensions.Optional[numpy.typing.NDArray] = None

Static depth image, which is used as depth image for all rgb images

update_global_with_depth_parameter: bool = True

Flag to update global depth parameters

cam_info: typing_extensions.Dict[str, typing_extensions.Any] = None

Camera config as dict

cam_intrinsic = None

Camera intrinsic parameters

color2depth_ratio: typing_extensions.Tuple[float, float] = (1.0, 1.0)

If the resolution of the depth image differs from the color image, we need to define the factor for (x, y).

viewpoint_cam_to_world = None

Camera to world transformation