Overview
The VOIDQuadmaskPreprocess node prepares a mask for VOID inpainting by converting it into a special four-level “quadmask.” It takes an input mask, optionally dilates the primary region, then quantizes the mask values into four distinct levels representing different semantic regions (primary object, overlap, affected area, and background). Finally, it inverts and normalizes the mask so the output values are in the range [0, 1], where 1.0 indicates the area to remove and 0.0 indicates the area to keep.Inputs
Outputs
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub
Source fingerprint (SHA-256):
a972ff74686baaa3a3659a74517d3b6f7335e8f62d09be9329d0e71d1fad137a