COCOSegToSegMask

class hybrid_learning.datasets.custom.coco.keypoints_dataset.COCOSegToSegMask(img_size=None, coco_handle=None, merge_masks=True, include_crowds=True)

Bases: _ToSegMask

Turn COCO instance segmentation annotations into a single binary segmentation mask. The reduction is the trivial logical AND, as masks are assumed to be binary. Output size: (masks, height, width).

Public Data Attributes:

Inherited from : py: class:_ToSegMask

settings

Settings to reproduce the instance.

Inherited from : py: class:Transform

IDENTITY_CLASS	The identity class or classes for composition / addition.
settings	Settings to reproduce the instance.

Public Methods:

to_masks(anns)

Turn annotations into list of boolean masks.

Inherited from : py: class:_ToSegMask

to_masks(anns)	Turn annotations into list of boolean masks.
apply_to(anns)	Turn annotations into a single binary float tensor.

Inherited from : py: class:Transform

apply_to(anns)

Turn annotations into a single binary float tensor.

Special Methods:

Inherited from : py: class:_ToSegMask

__init__([img_size, coco_handle, ...])

Init.

Inherited from : py: class:Transform

__repr__()	Return repr(self).
__eq__(other)	Return self==value.
__copy__()	Return a shallow copy of self using settings.
__add__(other)	Return a flat composition of self with other.
__radd__(other)	Return a flat composition of other and self.
__call__(*inps)	General call to the transformation.

---

Parameters

• img_size (Tuple[int, int]) –

• coco_handle (COCO) –

• merge_masks (bool) –

• include_crowds (bool) –

to_masks(anns)

Turn annotations into list of boolean masks.

Parameters

anns (Sequence[Dict[str, Any]]) –

Return type

List[ndarray]