LayerMask: MaskByDifferent

Documentation

• Class name: LayerMask: MaskByDifferent
• Category: 😺dzNodes/LayerMask
• Output node: False

This node is designed to create a mask based on differences between two images, enabling the identification and isolation of unique elements. It facilitates dynamic segmentation by analyzing and contrasting the visual characteristics present in each layer.

Input types

Required

• image_i
  • The first and second images to be compared, serving as reference points for identifying differences. These images are used to detect variations and unique features, enabling the creation of a mask that highlights these differences.
  • Comfy dtype: IMAGE
  • Python dtype: List[torch.Tensor]
• gain
  • Adjusts the sensitivity of the difference detection, influencing the intensity of the identified changes.
  • Comfy dtype: FLOAT
  • Python dtype: float
• fix_gap
  • Defines the tolerance for gap fixing in the generated mask, helping to smooth and refine the mask's edges.
  • Comfy dtype: INT
  • Python dtype: int
• fix_threshold
  • Sets the threshold for fixing gaps within the mask, further enhancing mask accuracy by refining edge details.
  • Comfy dtype: FLOAT
  • Python dtype: float
• main_subject_detect
  • A boolean flag that, when enabled, focuses the difference detection on the main subject of the images, potentially improving the relevance of the generated mask.
  • Comfy dtype: BOOLEAN
  • Python dtype: bool

Optional

Output types

• mask
  • Comfy dtype: MASK
  • The resulting mask that highlights the differences between the two input images, isolating unique elements and features.
  • Python dtype: torch.Tensor

Usage tips

• Infra type: CPU
• Common nodes: unknown

Source code

class MaskByDifferent:

    def __init__(self):
        pass

    @classmethod
    def INPUT_TYPES(self):

        return {
            "required": {
                "image_1": ("IMAGE", ),  #
                "image_2": ("IMAGE",),  #
                "gain": ("FLOAT", {"default": 1.5, "min": 0.1, "max": 100, "step": 0.1}),
                "fix_gap": ("INT", {"default": 4, "min": 0, "max": 32, "step": 1}),
                "fix_threshold": ("FLOAT", {"default": 0.75, "min": 0.01, "max": 0.99, "step": 0.01}),
                "main_subject_detect": ("BOOLEAN", {"default": False}),
            },
            "optional": {
            }
        }

    RETURN_TYPES = ( "MASK",)
    RETURN_NAMES = ("mask",)
    FUNCTION = 'mask_by_different'
    CATEGORY = '😺dzNodes/LayerMask'

    def mask_by_different(self, image_1, image_2, gain, fix_gap, fix_threshold, main_subject_detect):

        image1s = []
        image2s = []
        ret_masks = []
        for i in image_1:
            image1s.append(torch.unsqueeze(i, 0))
        for i in image_2:
            image2s.append(torch.unsqueeze(i, 0))
        max_batch = max(len(image1s), len(image2s))
        blank_mask = image2mask(Image.new('L', size=tensor2pil(image1s[0]).size, color='black'))
        if tensor2pil(image1s[0]).size != tensor2pil(image2s[0]).size:
            log(f"Error: {NODE_NAME} skipped, because the image size is not match.", message_type='error')
            return (torch.cat([blank_mask], dim=0))
        for i in range(max_batch):
            t1 = image1s[i] if i < len(image1s) else image1s[-1]
            t2 = image2s[i] if i < len(image2s) else image2s[-1]
            t1 = pil2tensor(tensor2pil(t1).convert('RGB'))
            t2 = pil2tensor(tensor2pil(t2).convert('RGB'))
            t = torch.abs(t1 - t2) * gain
            _mask = mask_fix(t, 1, fix_gap, fix_threshold, fix_threshold)
            _mask = tensor2pil(_mask)
            if main_subject_detect:
                subject_mask1 = RMBG(tensor2pil(t1))
                subject_mask2 = RMBG(tensor2pil(t2))
                subject_mask = chop_image(subject_mask1, subject_mask2, blend_mode='add', opacity=100)
                grow = (subject_mask.width + subject_mask.height) // 100
                subject_mask = mask2image(expand_mask(image2mask(subject_mask), grow * 2, grow))
                black = Image.new('L', size=_mask.size, color='black')
                white = Image.new('L', size=_mask.size, color='white')
                black.paste(_mask, mask=subject_mask.convert('L'))
                black.paste(white, mask=subject_mask1.convert('L'))
                black.paste(white, mask=subject_mask2.convert('L'))
                _mask = black

            ret_masks.append(image2mask(_mask))

        log(f"{NODE_NAME} Processed {len(ret_masks)} mask(s).", message_type='finish')
        return (torch.cat(ret_masks, dim=0),)