nerfbaselines.types¶
- class nerfbaselines.types.Dataset[source]¶
Bases:
_IncompleteDataset
- image_paths: List[str]¶
- image_paths_root: str¶
- images: ndarray | List[ndarray]¶
- images_points3D_indices: List[ndarray] | None¶
- metadata: Dict¶
- points3D_rgb: ndarray | None¶
- points3D_xyz: ndarray | None¶
- sampling_mask_paths: List[str] | None¶
- sampling_mask_paths_root: str | None¶
- sampling_masks: ndarray | List[ndarray] | None¶
- class nerfbaselines.types.DatasetSpecMetadata[source]¶
Bases:
TypedDict
- default_metric: str¶
- description: str¶
- id: str¶
- link: str¶
- metrics: List[str]¶
- name: str¶
- paper_authors: List[str]¶
- paper_link: str¶
- paper_title: str¶
- scenes: List[Dict[str, str]]¶
- class nerfbaselines.types.EvaluationProtocol(*args, **kwargs)[source]¶
Bases:
Protocol
- evaluate(predictions: Iterable[RenderOutput], dataset: Dataset) Iterable[Dict[str, float | int]] [source]¶
- render(method: Method, dataset: Dataset) Iterable[RenderOutput] [source]¶
- class nerfbaselines.types.GenericCameras(*args, **kwargs)[source]¶
Bases:
Protocol
[TTensor_co
]- apply(fn: Callable[[TTensor_co, str], TTensor]) GenericCameras[TTensor] [source]¶
- property camera_types: TTensor_co¶
Camera types, [N]
- property distortion_parameters: TTensor_co¶
Distortion parameters, [N, num_params]
- property image_sizes: TTensor_co¶
Image sizes, [N, 2]
- property intrinsics: TTensor_co¶
Intrinsics, [N, (fx,fy,cx,cy)]
- property metadata: TTensor_co | None¶
Metadata, [N, …]
- property nears_fars: TTensor_co | None¶
Near and far planes, [N, 2]
- property poses: TTensor_co¶
Camera-to-world matrices, [N, (R, t)]
- class nerfbaselines.types.GenericCamerasImpl(poses: +TTensor_co, intrinsics: +TTensor_co, camera_types: +TTensor_co, distortion_parameters: +TTensor_co, image_sizes: +TTensor_co, nears_fars: Optional[+TTensor_co], metadata: Optional[+TTensor_co] = None)[source]¶
Bases:
Generic
[TTensor_co
]- apply(fn: Callable[[TTensor_co, str], TTensor]) GenericCamerasImpl[TTensor] [source]¶
- camera_types: TTensor_co¶
- distortion_parameters: TTensor_co¶
- image_sizes: TTensor_co¶
- intrinsics: TTensor_co¶
- metadata: TTensor_co | None = None¶
- nears_fars: TTensor_co | None¶
- poses: TTensor_co¶
- class nerfbaselines.types.ImageSetInterpolationSource[source]¶
Bases:
TypedDict
- default_appearance: NotRequired[TrajectoryFrameAppearance | None]¶
- default_fov: float¶
- default_transition_duration: float¶
- interpolation: Literal['none']¶
- keyframes: List[TrajectoryKeyframe]¶
- type: Literal['interpolation']¶
- class nerfbaselines.types.KochanekBartelsInterpolationSource[source]¶
Bases:
TypedDict
- default_appearance: NotRequired[TrajectoryFrameAppearance | None]¶
- default_fov: float¶
- default_transition_duration: float¶
- interpolation: Literal['kochanek-bartels']¶
- is_cycle: bool¶
- keyframes: List[TrajectoryKeyframe]¶
- tension: float¶
- type: Literal['interpolation']¶
- class nerfbaselines.types.Logger(*args, **kwargs)[source]¶
Bases:
Protocol
- add_embedding(tag: str, embeddings: ndarray, step: int, *, images: List[ndarray] | None = None, labels: None | List[Dict[str, str]] | List[str] = None) None [source]¶
- add_event(step: int) ContextManager[LoggerEvent] [source]¶
- class nerfbaselines.types.LoggerEvent(*args, **kwargs)[source]¶
Bases:
Protocol
- add_embedding(tag: str, embeddings: ndarray, *, images: List[ndarray] | None = None, labels: None | List[Dict[str, str]] | List[str] = None) None [source]¶
- add_image(tag: str, image: ndarray, display_name: str | None = None, description: str | None = None, **kwargs) None [source]¶
- class nerfbaselines.types.Trajectory[source]¶
Bases:
TypedDict
- appearances: NotRequired[List[TrajectoryFrameAppearance]]¶
- camera_model: Literal['pinhole', 'opencv', 'opencv_fisheye', 'full_opencv']¶
- fps: float¶
- frames: List[TrajectoryFrame]¶
- image_size: Tuple[int, int]¶
- source: NotRequired[ImageSetInterpolationSource | KochanekBartelsInterpolationSource | None]¶
- class nerfbaselines.types.TrajectoryFrame[source]¶
Bases:
TypedDict
- appearance_weights: NotRequired[ndarray]¶
- intrinsics: ndarray¶
- pose: ndarray¶
- class nerfbaselines.types.TrajectoryFrameAppearance[source]¶
Bases:
TypedDict
- embedding: ndarray | None¶
- embedding_train_index: int | None¶
- class nerfbaselines.types.TrajectoryKeyframe[source]¶
Bases:
TypedDict
- appearance: NotRequired[TrajectoryFrameAppearance]¶
- fov: float | None¶
- pose: ndarray¶
- transition_duration: NotRequired[float | None]¶
- class nerfbaselines.types.UnloadedDataset[source]¶
Bases:
_IncompleteDataset
- image_paths: List[str]¶
- image_paths_root: str¶
- images: NotRequired[ndarray | List[ndarray] | None]¶
- images_points3D_indices: List[ndarray] | None¶
- metadata: Dict¶
- points3D_rgb: ndarray | None¶
- points3D_xyz: ndarray | None¶
- sampling_mask_paths: List[str] | None¶
- sampling_mask_paths_root: str | None¶
- sampling_masks: ndarray | List[ndarray] | None¶
- nerfbaselines.types.camera_model_from_int(i: int) Literal['pinhole', 'opencv', 'opencv_fisheye', 'full_opencv'] [source]¶