Nodes/ComfyUI-RemoveBackground_SET/Remove background (full)
ComfyUI Node

Remove background (full)

Removes or replaces the background from the input image. Gives more options and also generates masks and other stuff.

By set-soft·Created 12 months ago·Updated 8 months ago· 18
Remove background (full)
  • model
  • images
  • depths
  • background
  • images
  • masks
  • depths
  • edges
width1024
height1024
upscale_methodbicubic
blur_size90
blur_size_two6
fill_colorfalse
color#000000
mask_threshold0.000
batch_size1
out_dtypeAUTO
CategoryRemBG_SET/Advanced

Inputs (14)

NameTypeDefaultDescription
modelSET_REMBGThe remove background model from `Load RemBG model by file` or any of the `Load XXXXXX model by name` nodes
imagesIMAGEOne or more images to process, will be scaled to a size that is good for the model.
widthINT10240–16384The width of the pre-processing image, does not affect the final output image size
heightINT10240–16384The height of the pre-processing image, does not affect the final output image size
upscale_methodCOMBObicubicInterpolation method for pre-processing image and post-processing mask
blur_sizeINT901–255Diameter for the coarse gaussian blur used for the `Approximate Fast Foreground Colour Estimation`
blur_size_twoINT61–255Diameter for the fine gaussian blur (see `blur_size`)
fill_colorBOOLEANfalseWhen enabled and no background is provided we fill the background using a color.
colorSTRING#000000Color for fill. Can be an hexadecimal (#RRGGBB). Can be comma separated RGB values in [0-255] or [0-1.0] range. Also `rgb(R,G,B)`, `hsv(H,S,V)` and `hsl(H,S,L) are supported. 'In addition around 140 color names are recognized
mask_thresholdFLOAT0.0000–1Most models generates masks that contain a value from 0 to 1, but can be any value in between. Matte models can estimate transparency using it. If you need to make the mask 0 or 1, but nothing in between, you can provide a threshold here. Values above it will become 1 and the rest 0.
batch_sizeINT11–256How many images to process at once
depthsoptMASKFor models that starts with a depth map
backgroundoptIMAGEImage to use as background
out_dtypeoptCOMBOAUTOData type used for the outputs. `AUTO` means the same as the input.Using `float16` can help when processing videos.

Outputs (4)

NameTypeDescription
imagesIMAGEThe images with the background removed (transparent) or replaced by the background image
masksMASKThe estimated masks, where a higher value means the model estimates it belongs to the foreground with more confidence.
depthsMASKThe estimated depth map. Either from the `depths` input or computed. Note this applies only to PDFNet. This is the map generated by `Depth Anything V2`
edgesMASKThe estimated edges. This is only generated by the DiffDIS model.