ComfyUI Extension: comfyui's gaffer(ComfyUI native implementation of IC-Light. )

comfyui's gaffer(ComfyUI native implementation of IC-Light. ) Introduction

ComfyUI-Gaffer is an extension designed to enhance the capabilities of ComfyUI by integrating advanced lighting models, specifically the IC-Light models. This extension allows AI artists to manipulate the illumination of images in a highly controlled and consistent manner. By using ComfyUI-Gaffer, you can achieve professional-grade lighting effects in your AI-generated images, making them more realistic and visually appealing. Whether you are working on text-conditioned relighting or background-conditioned models, ComfyUI-Gaffer provides the tools you need to bring your artistic vision to life.

How comfyui's gaffer(ComfyUI native implementation of IC-Light. ) Works

ComfyUI-Gaffer works by integrating IC-Light models into the ComfyUI environment. These models are designed to manipulate the lighting of images based on various conditions such as text prompts, foreground images, and background images. Think of it as a virtual lighting studio where you can control the direction, intensity, and color of the light to achieve the desired effect. The extension uses nodes to load and apply these models, allowing you to customize the lighting conditions for each image. By adjusting parameters like light position and strength, you can create a wide range of lighting effects, from subtle ambient light to dramatic spotlighting.

comfyui's gaffer(ComfyUI native implementation of IC-Light. ) Features

Load ICLight Model Node

This node allows you to load an IC-Light model into your workflow. You can specify the model name, and the node will output the model information needed for further processing.

Input:

• iclight_name: The name of the IC-Light model to load. Output:
• iclight: IC-Light model information. Load ICLight Model

Apply ICLight Node

This node applies the loaded IC-Light model to your image. You can customize various parameters to control the lighting effect.

Input:

• model: Base model.
• vae: VAE.
• iclight: IC-Light model information.
• position: Position prompts.
• negative: Negative prompts.
• fg_pixels: Foreground image (background should be removed and set to grayscale).
• bg_pixels: Background image (optional, required for Background-Conditioned Model).
• multiplier: Strength of the IC-Light model. Output:
• model: Model with IC-Light applied.
• position: Position prompts.
• negative: Negative prompts.
• empty_latent: Empty Latent Image (dimensions consistent with the foreground image). Apply ICLight

Simple Light Source Node

This node allows you to create a simple light source for lighting preferences. You can specify the light position, strength, and colors to achieve different lighting effects.

Input:

• light_position: Position of the light (e.g., Left Light, Right Light).
• multiplier: Strength of the lighting preference.
• start_color: Start color of the lighting preference.
• end_color: End color of the lighting preference.
• width: Width of the lighting preference.
• height: Height of the lighting preference. Output:
• image: Lighting preference image. Simple Light Source Node

Calculate Normal Map Node

This node calculates a normal map from a sequence of input images. Normal maps are used to simulate lighting effects on surfaces.

Input:

• images: A sequence of input images.
• mask: Optional mask image to specify the computation region.
• sigma: Standard deviation for Gaussian blur.
• center_input_range: Range used to center the input images. Output:
• normal: Normal map. Calculate Normal Map

Gray Scaler Node

This node scales the image area to gray according to the provided mask, allowing you to control the overall brightness of the image.

Input:

• image: Transparent background image.
• mask: Mask indicating areas to be converted to gray.
• multiplier: Intensity of the gray conversion. Output:
• image: Gray background image. Gray Scaler Node

comfyui's gaffer(ComfyUI native implementation of IC-Light. ) Models

ComfyUI-Gaffer supports several IC-Light models, each designed for specific workflows:

• iclight_sd15_fc.safetensors: Use this model for foreground (FG) workflows.
• iclight_sd15_fcon.safetensors: Another model for FG workflows, trained with offset noise.
• iclight_sd15_fbc.safetensors: Use this model for background (BG) workflows. These models can be downloaded from Hugging Face and should be placed under ComfyUI/models/iclight.

Troubleshooting comfyui's gaffer(ComfyUI native implementation of IC-Light. )

Common Issues and Solutions

1. Model Not Loading Properly

• Ensure that the model files are placed in the correct directory (ComfyUI/models/iclight).
• Restart ComfyUI after placing the model files.

2. Lighting Effects Not Visible

• Check the parameters in the Apply ICLight Node to ensure they are set correctly.
• Make sure the foreground image has a removed background and is set to grayscale.

3. Errors During Workflow Execution

• Verify that all required nodes are installed, such as ComfyUI-KJNodes.
• Ensure that the input images and masks are correctly formatted and sized.

Frequently Asked Questions

• Q: Can I use my own models with ComfyUI-Gaffer?
• A: Currently, ComfyUI-Gaffer supports IC-Light models. Custom models are not supported.
• Q: How do I control the overall brightness of the image?
• A: Use the Gray Scaler Node to adjust the brightness by converting areas to gray.

Learn More about comfyui's gaffer(ComfyUI native implementation of IC-Light. )

For additional resources, tutorials, and community support, you can visit the following links:

• IC-Light GitHub Repository
• ComfyUI GitHub Repository
• ComfyUI-KJNodes GitHub Repository These resources provide comprehensive documentation and examples to help you get the most out of ComfyUI-Gaffer.