Guided Filter Alpha

Documentation

• Class name: GuidedFilterAlpha
• Category: image/filters
• Output node: False

The GuidedFilterAlpha node applies a guided filter to images using a specified alpha channel and filter parameters. This process enhances the image by smoothing while preserving edges, guided by the alpha channel to control the blending and detail levels.

Input types

Required

• images
  • The input images to be filtered, where the alpha channel guides the filtering process.
  • Comfy dtype: IMAGE
  • Python dtype: torch.Tensor
• alpha
  • The alpha channel used to guide the filtering process, influencing how the image details are blended and preserved.
  • Comfy dtype: IMAGE
  • Python dtype: torch.Tensor
• filter_radius
  • Specifies the radius of the filter, affecting the extent of smoothing and edge preservation.
  • Comfy dtype: INT
  • Python dtype: int
• sigma
  • Controls the degree of smoothing, with higher values resulting in more blur.
  • Comfy dtype: FLOAT
  • Python dtype: float

Output types

• image
  • Comfy dtype: IMAGE
  • The filtered images, where the guided filter has been applied according to the alpha channel and filter parameters.
  • Python dtype: torch.Tensor

Usage tips

• Infra type: GPU
• Common nodes: unknown

Source code

class GuidedFilterAlpha:
    def __init__(self):
        pass

    @classmethod
    def INPUT_TYPES(s):
        return {
            "required": {
                "images": ("IMAGE",),
                "alpha": ("IMAGE",),
                "filter_radius": ("INT", {
                    "default": 8,
                    "min": 1,
                    "max": 64,
                    "step": 1
                }),
                "sigma": ("FLOAT", {
                    "default": 0.1,
                    "min": 0.01,
                    "max": 1.0,
                    "step": 0.01
                }),
            },
        }

    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "guided_filter_alpha"

    CATEGORY = "image/filters"

    def guided_filter_alpha(self, images: torch.Tensor, alpha: torch.Tensor, filter_radius: int, sigma: float):

        d = filter_radius * 2 + 1
        s = sigma / 10

        i_dup = copy.deepcopy(images.cpu().numpy())
        a_dup = copy.deepcopy(alpha.cpu().numpy())

        for index, image in enumerate(i_dup):
            alpha_work = a_dup[index]
            i_dup[index] = guidedFilter(image, alpha_work, d, s)

        return (torch.from_numpy(i_dup),)