SAM Parameters Combine

Documentation

• Class name: SAM Parameters Combine
• Category: WAS Suite/Image/Masking
• Output node: False

This node combines the parameters of two SAM (Spatial Attention Model) operations, effectively merging their points and labels into a single set. It's designed to facilitate the application of multiple SAM operations on a single image or dataset by consolidating their parameters for streamlined processing.

Input types

Required

• sam_parameters_a
  • The first set of SAM parameters to be combined. It includes points and labels that specify regions of interest and their corresponding labels in an image.
  • Comfy dtype: SAM_PARAMETERS
  • Python dtype: Dict[str, np.ndarray]
• sam_parameters_b
  • The second set of SAM parameters to be combined. Similar to the first, it includes points and labels for regions of interest in an image.
  • Comfy dtype: SAM_PARAMETERS
  • Python dtype: Dict[str, np.ndarray]

Output types

• sam_parameters
  • Comfy dtype: SAM_PARAMETERS
  • The combined SAM parameters, including points and labels from both input sets, ready for use in subsequent SAM operations.
  • Python dtype: Dict[str, np.ndarray]

Usage tips

• Infra type: CPU
• Common nodes: unknown

Source code

class WAS_SAM_Combine_Parameters:
    def __init__(self):
        pass

    @classmethod
    def INPUT_TYPES(self):
        return {
            "required": {
                "sam_parameters_a": ("SAM_PARAMETERS",),
                "sam_parameters_b": ("SAM_PARAMETERS",),
            }
        }

    RETURN_TYPES = ("SAM_PARAMETERS",)
    FUNCTION = "sam_combine_parameters"

    CATEGORY = "WAS Suite/Image/Masking"

    def sam_combine_parameters(self, sam_parameters_a, sam_parameters_b):
        parameters = {
            "points": np.concatenate(
                (sam_parameters_a["points"],
                sam_parameters_b["points"]),
                axis=0
            ),
            "labels": np.concatenate(
                (sam_parameters_a["labels"],
                sam_parameters_b["labels"])
            )
        }

        return (parameters,)