`pixano_inference.pydantic.tasks.image.mask_generation`

Pydantic models for image mask generation task.

`ImageMaskGenerationInput`

Input for image mask generation.

Attributes:

Name	Type	Description
`image`	`str \| Path`	Image for image mask generation.
`image_embedding`	`NDArrayFloat \| LanceVector \| None`	Image embedding for the image mask generation.
`high_resolution_features`	`list[NDArrayFloat] \| list[LanceVector] \| None`	High resolution features for the image mask generation.
`points`	`list[list[list[int]]] \| None`	Points for the image mask generation. The first fimension is the number of prompts the second the number of points per mask and the third the coordinates of the points.
`labels`	`list[list[int]] \| None`	Labels for the image mask generation. The first fimension is the number of prompts, the second the number of labels per mask.
`boxes`	`list[list[int]] \| None`	Boxes for the image mask generation. The first fimension is the number of prompts, the second the coordinates of the boxes.
`num_multimask_outputs`	`int`	Number of masks to generate per prediction.
`multimask_output`	`bool`	Whether to generate multiple masks per prediction.
`return_image_embedding`	`bool`	Whether to return the image embeddings.

Output for image mask generation.

Attributes:

Name	Type	Description
`masks`	`list[list[CompressedRLE]]`	Generated masks. The first dimension is the number of predictions, the second the number of masks per prediction.
`scores`	`NDArrayFloat`	Scores of the masks. The first dimension is the number of predictions, the second the number of masks per prediction.
`image_embedding`	`NDArrayFloat \| None`	Image embeddings.
`high_resolution_features`	`list[NDArrayFloat] \| None`	High resolution features.

Request for image mask generation.

Convert the request to the input.

Source code in pixano_inference/pydantic/tasks/image/mask_generation.py

def to_input(self) -> ImageMaskGenerationInput:
    """Convert the request to the input."""
    return self.to_base_model(ImageMaskGenerationInput)

Response for image mask generation.

Attributes:

Name	Type	Description
`data`	`ImageMaskGenerationOutput`	Output of the generation.