pyflink.multimodal.expression.ImageExpressionAccessor.remove_background

ImageExpressionAccessor.remove_background(*, alpha_matting: bool = False, alpha_matting_foreground_threshold: int = 240, alpha_matting_background_threshold: int = 10, alpha_matting_erode_size: int = 10, bgcolor: Optional[Union[List[int], Tuple[int, int, int, int]]] = None, model_sharing: Optional[Literal['process', 'shared']] = None, concurrency: Optional[int] = None, batch_size: Optional[int] = None, num_gpus: Optional[float] = None, gpu_type: Optional[str] = None) → pyflink.table.expression.Expression

Remove image backgrounds and optionally composite a new background.

Equivalent to image_remove_background(). See that function for full parameter details.

Parameters

• alpha_matting – Whether to refine the foreground mask with alpha matting.

• alpha_matting_foreground_threshold – Foreground threshold for alpha matting.

• alpha_matting_background_threshold – Background threshold for alpha matting.

• alpha_matting_erode_size – Erosion size for alpha matting.

• bgcolor – Optional replacement background color.

• model_sharing – Optional model sharing mode.

• concurrency – Optional execution concurrency for this operation. None uses the framework default.

• batch_size – Optional inference batch size.

• num_gpus – Optional number of GPUs requested for model inference.

• gpu_type – GPU type requested for model inference. The DataFrame UDF runtime requires this when num_gpus is set.

Returns

A DataFrame expression that produces foreground IMAGE values, optionally composited with the requested background.