SeedVR2PostProcessing - ComfyUI Built-in Node Documentation

This node aligns the generated image with the original resized image and applies optional color correction. It takes the output from a SeedVR2 upscaling process and adjusts it to match the colors and dimensions of the original reference image.

Inputs

Parameter	Description	Data Type	Required	Range
`images`	The generated image to process.	IMAGE	Yes	-
`original_resized_images`	The original resized image before pre-processing, used as reference.	IMAGE	Yes	-
`color_correction_method`	Method to match the generated image colors to the original image. lab: transfer color in CIELAB space, preserving detail (most faithful). wavelet: transfer low-frequency color, keeping upscaled high-frequency detail. adain: match per-channel mean/std (fastest, global tint). none: skip color transfer (geometry alignment only). (default: “lab”)	COMBO	Yes	`"lab"` `"wavelet"` `"adain"` `"none"`

Note: Both inputs may be 4-D (batch, height, width, channels) or 5-D (batch, frames, height, width, channels) tensors. The node crops both to the smallest batch, frame count, height, and width, so they do not need to match exactly. Output height and width are rounded down to even numbers. If the reference image has an alpha channel (4 channels), that alpha channel is preserved and applied to the output.

Outputs

Output Name	Description	Data Type
`images`	The aligned, color-corrected image.	IMAGE

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): 00a3a3ef06edc7e0eca8f67a96095920a3e0e885dac3fb676d081e4c4c30bec5

​Inputs

​Outputs

Inputs

Outputs