Box Detections

class py123d.datatypes.BoxDetectionsSE3[source]

Container for a list of SE3 box detections.

Provides indexed access, iteration, lookup by track token, and a 2D occupancy map.

Public Data Attributes:

box_detections

List of BoxDetectionSE3 instances.

timestamp

The Timestamp of the box detections.

metadata

The metadata for the box detections modality.

Inherited from BaseModality

timestamp

Returns the timestamp associated with this modality data, if available.

metadata

Returns the metadata associated with this modality data.

modality_type

Convenience property to access the modality type directly from the modality data.

modality_id

Convenience property to access the modality id directly from the modality data.

modality_key

Convenience property to access the modality key directly from the modality data.

Public Methods:

get_detection_by_track_token(track_token)

Retrieve a box detection by its track token.

occupancy_map_2d

The OccupancyMap2D representing the 2D occupancy of all box detections.


property box_detections: List[BoxDetectionSE3]

List of BoxDetectionSE3 instances.

property timestamp: Timestamp

The Timestamp of the box detections.

property metadata: BoxDetectionsSE3Metadata

The metadata for the box detections modality.

get_detection_by_track_token(track_token)[source]

Retrieve a box detection by its track token.

Parameters:

track_token (str) – The track token of the box detection.

Return type:

Optional[BoxDetectionSE3]

Returns:

The BoxDetectionSE3 with the given track token, or None if not found.

property occupancy_map_2d: OccupancyMap2D

The OccupancyMap2D representing the 2D occupancy of all box detections.

property modality_id: str | SerialIntEnum | None

Convenience property to access the modality id directly from the modality data.

property modality_key: str

Convenience property to access the modality key directly from the modality data.

property modality_type: ModalityType

Convenience property to access the modality type directly from the modality data.

class py123d.datatypes.BoxDetectionSE3[source]

Detected, tracked, and oriented bounding box 3D space.

Public Data Attributes:

attributes

The BoxDetectionAttributes of the detection.

bounding_box_se3

The BoundingBoxSE3 of the detection.

bounding_box_se2

The SE2 projection BoundingBoxSE2 of the SE3 bounding box.

velocity_3d

The Vector3D representing the velocity in global frame.

velocity_2d

The 2D projection Vector2D of the 3D velocity vector in global frame.

center_se3

The PoseSE3 representing the center of the bounding box.

center_se2

The PoseSE2 representing the center of the SE2 bounding box.

box_detection_se2

The BoxDetectionSE2 projection of this SE3 box detection.

shapely_polygon

The shapely polygon of the bounding box in 2D space.


property attributes: BoxDetectionAttributes

The BoxDetectionAttributes of the detection.

property bounding_box_se3: BoundingBoxSE3

The BoundingBoxSE3 of the detection.

property bounding_box_se2: BoundingBoxSE2

The SE2 projection BoundingBoxSE2 of the SE3 bounding box.

property velocity_3d: Vector3D | None

The Vector3D representing the velocity in global frame.

property velocity_2d: Vector2D | None

The 2D projection Vector2D of the 3D velocity vector in global frame.

property center_se3: PoseSE3

The PoseSE3 representing the center of the bounding box.

property center_se2: PoseSE2

The PoseSE2 representing the center of the SE2 bounding box.

property box_detection_se2: BoxDetectionSE2

The BoxDetectionSE2 projection of this SE3 box detection.

property shapely_polygon: Polygon

The shapely polygon of the bounding box in 2D space.

class py123d.datatypes.BoxDetectionAttributes[source]

Stores data about the box detection, including its label, track token, and number of Lidar points.

Public Data Attributes:

label

The BoxDetectionLabel, from the original dataset's label set.

track_token

The unique track token of the detection, consistent across frames.

num_lidar_points

Optionally, the number of Lidar points associated with the detection.

default_label

The unified DefaultBoxDetectionLabel corresponding to the detection's label.


property label: BoxDetectionLabel

The BoxDetectionLabel, from the original dataset’s label set.

property track_token: str

The unique track token of the detection, consistent across frames.

property num_lidar_points: int | None

Optionally, the number of Lidar points associated with the detection.

property default_label: DefaultBoxDetectionLabel

The unified DefaultBoxDetectionLabel corresponding to the detection’s label.

class py123d.datatypes.BoxDetectionsSE2[source]

Container for a list of SE2 box detections.

Provides indexed access, iteration, lookup by track token, and a 2D occupancy map.

Public Data Attributes:

box_detections

List of BoxDetectionSE2 instances.

timestamp

The Timestamp of the box detections.

Public Methods:

get_detection_by_track_token(track_token)

Retrieve a box detection by its track token.

occupancy_map_2d

The OccupancyMap2D representing the 2D occupancy of all box detections.


property box_detections: List[BoxDetectionSE2]

List of BoxDetectionSE2 instances.

property timestamp: Timestamp

The Timestamp of the box detections.

get_detection_by_track_token(track_token)[source]

Retrieve a box detection by its track token.

Parameters:

track_token (str) – The track token of the box detection.

Return type:

Optional[BoxDetectionSE2]

Returns:

The BoxDetectionSE2 with the given track token, or None if not found.

property occupancy_map_2d: OccupancyMap2D

The OccupancyMap2D representing the 2D occupancy of all box detections.

class py123d.datatypes.BoxDetectionSE2[source]

Detected, tracked, and oriented bounding box 2D space.

Public Data Attributes:

attributes

The BoxDetectionAttributes of the detection.

bounding_box_se2

The BoundingBoxSE2 of the detection.

velocity_2d

The Vector2D representing the velocity in global frame.

center_se2

The PoseSE2 representing the center of the bounding box.

shapely_polygon

The shapely polygon of the bounding box in 2D space.

box_detection_se2

Returns self to maintain interface consistency.


property attributes: BoxDetectionAttributes

The BoxDetectionAttributes of the detection.

property bounding_box_se2: BoundingBoxSE2

The BoundingBoxSE2 of the detection.

property velocity_2d: Vector2D | None

The Vector2D representing the velocity in global frame.

property center_se2: PoseSE2

The PoseSE2 representing the center of the bounding box.

property shapely_polygon: Polygon

The shapely polygon of the bounding box in 2D space.

property box_detection_se2: BoxDetectionSE2

Returns self to maintain interface consistency.