StandardFrameData¶

class standard_e2e.data_structures.StandardFrameData(**data)[source]¶

Bases: BaseModel

Represents a single frame data in intermediate standardized format: Raw frame data -> StandardFrameData -> TransformedFrameData.

Parameters:

timestamp (float)
frame_id (int)
segment_id (str)
dataset_name (str)
split (str)
global_position (Trajectory | None)
intent (Intent | None)
cameras (dict[CameraDirection, CameraData])
lidar (LidarData | None)
future_states (Trajectory | None)
past_states (Trajectory | None)
hd_map (HDMap | None)
frame_detections_3d (FrameDetections3D | None)
aux_data (Dict[str, Any] | None)
extra_index_data (Dict[str, Any] | None)

timestamp¶

Timestamp of the frame in seconds.

Type:: float

frame_id¶

Unique identifier of the frame within a sequence.

Type:: int

segment_id¶

Unique identifier of the segment this frame belongs to.

Type:: str

dataset_name¶

Name of the dataset this frame belongs to.

Type:: str

split¶

Dataset split (e.g., “train”, “val”, “test”).

Type:: str

global_position¶

Pose for the ego entity at this frame in global coordinates.

Type:: Optional[Trajectory]

intent¶

Predicted or annotated intent associated with the frame.

Type:: Optional[Intent]

cameras¶

Camera data keyed by camera direction.

Type:: dict[CameraDirection, CameraData]

lidar¶

LiDAR data for the frame, if available.

Type:: Optional[LidarData]

future_states¶

Future trajectory states relative to this frame.

Type:: Optional[Trajectory]

past_states¶

Past trajectory states leading up to this frame.

Type:: Optional[Trajectory]

hd_map¶

HD map snapshot in vehicle frame at this frame’s timestamp; consumed by HDMapBEVAdapter (and future vector adapters). In-memory only — not persisted to .npz.

Type:: Optional[HDMap]

frame_detections_3d¶

3D detections present in the frame.

Type:: Optional[FrameDetections3D]

aux_data¶

Additional auxiliary data.

Type:: Optional[Dict[str, Any]]

extra_index_data¶

Extra indexing or lookup data.

Type:: Optional[Dict[str, Any]]

aux_data: Dict[str, Any] | None¶

cameras: dict[CameraDirection, CameraData]¶

dataset_name: str¶

extra_index_data: Dict[str, Any] | None¶

frame_detections_3d: FrameDetections3D | None¶

frame_id: int¶

future_states: Trajectory | None¶

global_position: Trajectory | None¶

hd_map: HDMap | None¶

intent: Intent | None¶

lidar: LidarData | None¶

past_states: Trajectory | None¶

segment_id: str¶

split: str¶

timestamp: float¶