fiftyone.core.labels.Segmentation

class documentation

class Segmentation(_HasID, _HasMedia, Label): (source)

A semantic segmentation for an image.

Provide either the mask or mask_path argument to define the segmentation.

Parameters
mask	a numpy array with integer values encoding the semantic labels
mask_path	the absolute path to the segmentation image on disk

Method	`export_mask`	Exports this instance's mask to the given path.
Method	`get_mask`	Returns the segmentation mask for this instance.
Method	`import_mask`	Imports this instance's mask from disk to its `mask` attribute.
Method	`to_detections`	Returns a `Detections` representation of this instance with instance masks populated.
Method	`to_polylines`	Returns a `Polylines` representation of this instance.
Method	`transform_mask`	Transforms this instance's mask according to the provided targets map.
Instance Variable	`mask`	Undocumented
Instance Variable	`mask_path`	Undocumented
Property	`has_mask`	Whether this instance has a mask.
Constant	`_MEDIA_FIELD`	Undocumented

Inherited from _HasID:

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

Inherited from Label (via _HasID, _HasMedia):

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.

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 segmentation 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_detections(self, mask_targets=None, mask_types='stuff'): (source) ¶

Returns a Detections representation of this instance with instance masks populated.

Each "stuff" class will be converted to a single Detection whose instance mask spans all region(s) of the class.

Each "thing" class will result in one Detection instance per connected region of that class in the segmentation.

Parameters
mask_targets:`None`	a dict mapping integer pixel values (2D masks) or RGB hex strings (3D masks) to label strings defining which classes to generate detections for. If omitted, all labels are assigned to their pixel values
mask_types:"stuff"	whether the classes are `"stuff"` (amorphous regions of pixels) or `"thing"` (connected regions, each representing an instance of the thing). Can be any of the following: `"stuff"` if all classes are stuff classes `"thing"` if all classes are thing classes a dict mapping pixel values (2D masks) or RGB hex strings (3D masks) to `"stuff"` or `"thing"` for each class
Returns
a `Detections`

def to_polylines(self, mask_targets=None, mask_types='stuff', tolerance=2): (source) ¶

Returns a Polylines representation of this instance.

Each "stuff" class will be converted to a single Polyline that may contain multiple disjoint shapes capturing the class.

Each "thing" class will result in one Polyline instance per connected region of that class.

Parameters
mask_targets:`None`	a dict mapping integer pixel values (2D masks) or RGB hex strings (3D masks) to label strings defining which classes to generate detections for. If omitted, all labels are assigned to their pixel values
mask_types:"stuff"	whether the classes are `"stuff"` (amorphous regions of pixels) or `"thing"` (connected regions, each representing an instance of the thing). Can be any of the following: `"stuff"` if all classes are stuff classes `"thing"` if all classes are thing classes a dict mapping pixel values (2D masks) or RGB hex strings (3D masks) to `"stuff"` or `"thing"` for each class
tolerance:2	a tolerance, in pixels, when generating approximate polylines for each region. Typical values are 1-3 pixels
Returns
a `Polylines`

def transform_mask(self, targets_map, outpath=None, update=False): (source) ¶

Transforms this instance's mask according to the provided targets map.

This method can be used to transform between grayscale and RGB masks, or it can be used to edit the pixel values or colors of a mask without changing the number of channels.

Note that any pixel values not in targets_map will be zero in the transformed mask.

Parameters
targets_map	a dict mapping existing pixel values (2D masks) or RGB hex strings (3D masks) to new pixel values or RGB hex strings. You may convert between grayscale and RGB using this argument
outpath:`None`	an optional path to write the transformed mask on disk
update:`False`	whether to save the transformed mask on this instance
Returns
the transformed mask

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'