fiftyone.core.labels.Detection

class documentation

class Detection(_HasAttributesDict, _HasID, _HasMedia, Label): (source)

An object detection.

Parameters
label	the label string
bounding_box	a list of relative bounding box coordinates in `[0, 1]` in the following format: [<top-left-x>, <top-left-y>, <width>, <height>]
mask	an instance segmentation mask for the detection within its bounding box, which should be a 2D binary or 0/1 integer numpy array
mask_path	the absolute path to the instance segmentation image on disk, which should be a single-channel PNG image where any non-zero values represent the instance's extent
confidence	a confidence in `[0, 1]` for the detection
index	an index for the object
attributes	a dict mapping attribute names to `Attribute` instances

Class Method	`from_mask`	Creates a `Detection` instance with its `mask` attribute populated from the given full image mask.
Method	`export_mask`	Exports this instance's mask to the given path.
Method	`get_mask`	Returns the detection mask for this instance.
Method	`import_mask`	Imports this instance's mask from disk to its `mask` attribute.
Method	`to_polyline`	Returns a `Polyline` representation of this instance.
Method	`to_segmentation`	Returns a `Segmentation` representation of this instance.
Method	`to_shapely`	Returns a Shapely representation of this instance.
Class Variable	`bounding_box`	Undocumented
Class Variable	`confidence`	Undocumented
Class Variable	`index`	Undocumented
Class Variable	`label`	Undocumented
Instance Variable	`mask`	Undocumented
Instance Variable	`mask_path`	Undocumented
Property	`has_mask`	Whether this instance has a mask.
Constant	`_MEDIA_FIELD`	Undocumented

Inherited from _HasAttributesDict:

Method	`delete_attribute`	Deletes the attribute with the given name.
Method	`get_attribute_value`	Gets the value of the attribute with the given name.
Method	`has_attribute`	Determines whether the label has an attribute with the given name.
Method	`iter_attributes`	Returns an iterator over the custom attributes of the label.
Method	`set_attribute_value`	Sets the value of the attribute with the given name.

Inherited from _HasID (via _HasAttributesDict):

Class Variable	`tags`	Undocumented
Instance Variable	`id`	Undocumented
Method	`_id.setter`	Undocumented
Property	`_id`	Undocumented

@classmethod
def from_mask(cls, mask, label=None, **attributes): (source) ¶

Creates a Detection instance with its mask attribute populated from the given full image mask.

The instance mask for the object is extracted by computing the bounding rectangle of the non-zero values in the image mask.

Parameters
mask	a boolean or 0/1 numpy array
label:`None`	the label string
**attributes	additional attributes for the `Detection`
Returns
a `Detection`

def export_mask(self, outpath, update=False): (source) ¶

Exports this instance's mask to the given path.

Parameters
outpath	the path to write the mask
update:`False`	whether to clear this instance's `mask` attribute and set its `mask_path` attribute when exporting in-database segmentations

def get_mask(self): (source) ¶

Returns the detection mask for this instance.

Returns
a numpy array, or `None`

def import_mask(self, update=False): (source) ¶

Imports this instance's mask from disk to its mask attribute.

Parameters
update:`False`	whether to clear this instance's `mask_path` attribute after importing

def to_polyline(self, tolerance=2, filled=True): (source) ¶

Returns a Polyline representation of this instance.

If the detection has a mask, the returned polyline will trace the boundary of the mask; otherwise, the polyline will trace the bounding box itself.

Parameters
tolerance:2	a tolerance, in pixels, when generating an approximate polyline for the instance mask. Typical values are 1-3 pixels
filled:`True`	whether the polyline should be filled
Returns
a `Polyline`

def to_segmentation(self, mask=None, frame_size=None, target=255): (source) ¶

Returns a Segmentation representation of this instance.

The detection must have an instance mask, i.e., its mask attribute must be populated.

You must provide either mask or frame_size to use this method.

Parameters
mask:`None`	an optional numpy array to use as an initial mask to which to add this object
frame_size:`None`	the `(width, height)` of the segmentation mask to render. This parameter has no effect if a `mask` is provided
target:255	the pixel value or RGB hex string to use to render the object
Returns
a `Segmentation`

def to_shapely(self, frame_size=None): (source) ¶

Returns a Shapely representation of this instance.

Parameters
frame_size:`None`	the `(width, height)` of the image. If provided, the returned geometry will use absolute coordinates
Returns
a `shapely.geometry.polygon.Polygon`

bounding_box: None = (source) ¶

Undocumented

confidence: None = (source) ¶

Undocumented

index: None = (source) ¶

Undocumented

label: None = (source) ¶

Undocumented

mask: None = (source) ¶

Undocumented

mask_path: None = (source) ¶

Undocumented

@property
has_mask = (source) ¶

Whether this instance has a mask.

_MEDIA_FIELD: str = (source) ¶

overrides fiftyone.core.labels._HasMedia._MEDIA_FIELD

Undocumented

Value

'mask_path'