ComfyUI > Nodes > ComfyUI-LCS > LCS Color Anchor

ComfyUI Node: LCS Color Anchor

Class Name

LCSColorAnchor

Category
LCS/intervention
Author
facok (Account age: 1139days)
Extension
ComfyUI-LCS
Latest Updated
2026-05-06
Github Stars
0.1K

How to Install ComfyUI-LCS

Install this extension via the ComfyUI Manager by searching for ComfyUI-LCS
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-LCS in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

LCS Color Anchor Description

Corrects color drift in image generation, maintains color consistency, offers adaptive operation with four modes catering to different scenarios.

LCS Color Anchor:

The LCSColorAnchor node is designed to correct color drift during the sampling process in image generation. This node is particularly useful for maintaining color consistency and accuracy by anchoring local color relationships. It operates adaptively, meaning it derives its scheduling and filtering parameters from runtime signals such as the sigma schedule and local color statistics. The node offers four modes of operation: auto, smooth, reference, and self_anchor, each catering to different scenarios of color correction. The auto mode intelligently infers the best mode and intensity based on the connected inputs and detected drift signals. The smooth mode applies a bilateral filter to smooth out color discontinuities, which is especially useful for inpainting boundaries. The reference mode anchors the color relationships to those of a reference image, ensuring that the generated image maintains a similar color palette. Lastly, the self_anchor mode builds an internal color model during the warmup phase and uses it to correct any drift. This node is essential for artists who want to ensure their generated images have consistent and accurate color representation.

LCS Color Anchor Input Parameters:

mode

The mode parameter determines the operational mode of the LCSColorAnchor node. It can be set to auto, smooth, reference, or self_anchor. Each mode has a specific function: auto mode automatically selects the best mode based on input signals; smooth mode applies a bilateral filter to smooth color transitions; reference mode uses a reference image to anchor colors; and self_anchor mode builds an internal color model for drift correction. The choice of mode significantly impacts how the node corrects color drift, with auto mode being the most adaptive and user-friendly option.

intensity

The intensity parameter controls the strength of the color correction applied by the node. A higher intensity value results in more aggressive color correction, while a lower value applies a subtler adjustment. If the intensity is set below a certain threshold (1e-6), the node may not perform any correction. This parameter allows you to fine-tune the degree of color correction to match your specific needs, ensuring that the final image maintains the desired color fidelity.

reference_image

The reference_image parameter is used in reference mode to provide a target image whose color relationships will be used to anchor the generated image. This parameter is crucial for ensuring that the generated image closely matches the color palette of the reference image. If no reference image is provided in reference mode, the node will skip the correction process, as it relies on this input to function correctly.

vae

The vae parameter is required in reference mode to encode the reference image into latent color space (LCS). The Variational Autoencoder (VAE) helps in translating the reference image into a format that the node can use to compare and correct colors. Without a VAE, the node cannot process the reference image, and the color anchoring will not occur.

mask

The mask parameter is used to specify areas of the image where color correction should be applied. This is particularly useful in smooth mode, where you might want to target specific regions for smoothing color transitions. The mask allows for selective application of the node's effects, providing greater control over the final image output.

LCS Color Anchor Output Parameters:

NodeOutput

The NodeOutput parameter is the primary output of the LCSColorAnchor node. It contains the corrected image model after the color anchoring process has been applied. This output is crucial for ensuring that the generated image maintains consistent and accurate color representation, as per the selected mode and intensity settings. The NodeOutput reflects the node's ability to adaptively correct color drift, providing a final image that aligns with the user's artistic vision.

LCS Color Anchor Usage Tips:

  • Use auto mode for a hassle-free experience, as it automatically selects the best mode and intensity based on input signals and detected drift.
  • When working with a specific color palette, use reference mode with a well-chosen reference image to ensure color consistency.
  • Adjust the intensity parameter to control the strength of the color correction, starting with a moderate value and fine-tuning as needed.
  • Utilize the mask parameter in smooth mode to target specific areas for color smoothing, enhancing the overall image quality.

LCS Color Anchor Common Errors and Solutions:

[LCS Color Anchor] Reference mode requires vae and reference_image — skipping.

  • Explanation: This error occurs when the node is set to reference mode but either the VAE or the reference image is not provided.
  • Solution: Ensure that both the VAE and a valid reference image are supplied when using reference mode.

[LCS Color Anchor] Failed to encode reference image — skipping.

  • Explanation: This error indicates that the node was unable to encode the reference image into the latent color space, possibly due to an issue with the VAE or the image itself.
  • Solution: Verify that the VAE is functioning correctly and that the reference image is in a compatible format. Try using a different image if the problem persists.

LCS Color Anchor Related Nodes

Go back to the extension to check out more related nodes.
ComfyUI-LCS
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

LCS Color Anchor