SEGS to Mask Batch¶
Documentation¶
- Class name:
ImpactSEGSToMaskBatch
- Category:
ImpactPack/Util
- Output node:
False
This node is designed to convert a collection of segmentation data (SEGS) into a batch of masks. It processes the input segmentation data to generate a corresponding set of masks, which are then combined into a single batch. This operation facilitates the handling and manipulation of mask data at scale, streamlining workflows that involve the analysis or transformation of segmented images.
Input types¶
Required¶
segs
- The 'segs' input represents the segmentation data that will be converted into a batch of masks. This data is crucial for the node's operation as it forms the basis for the mask generation process.
- Comfy dtype:
SEGS
- Python dtype:
List[torch.Tensor]
Output types¶
mask
- Comfy dtype:
MASK
- The output is a batch of masks, where each mask corresponds to a segment from the input segmentation data. This batch format is useful for subsequent processing or analysis steps that require masks in a collective form.
- Python dtype:
torch.Tensor
- Comfy dtype:
Usage tips¶
- Infra type:
GPU
- Common nodes: unknown
Source code¶
class SEGSToMaskBatch:
@classmethod
def INPUT_TYPES(s):
return {"required": {
"segs": ("SEGS", ),
},
}
RETURN_TYPES = ("MASK",)
FUNCTION = "doit"
CATEGORY = "ImpactPack/Util"
def doit(self, segs):
masks = core.segs_to_masklist(segs)
masks = [utils.make_3d_mask(mask) for mask in masks]
mask_batch = torch.concat(masks)
return (mask_batch,)