ComfyUI Extension: zhangp365/ComfyUI-utils-nodes

zhangp365/ComfyUI-utils-nodes Introduction

ComfyUI-utils-nodes is an extension designed to enhance the functionality of ComfyUI by providing a variety of utility nodes. These nodes are tailored to improve performance, add new features, and offer more control over image processing tasks. Whether you're an AI artist looking to streamline your workflow or seeking to add more sophisticated image manipulation capabilities, ComfyUI-utils-nodes can help you achieve your goals more efficiently.

How zhangp365/ComfyUI-utils-nodes Works

ComfyUI-utils-nodes works by extending the existing nodes in ComfyUI with additional features and optimizations. For example, some nodes are modified to include switches that allow you to enable or disable certain functionalities, which can save processing time and resources. Other nodes are optimized to handle large directories of images more efficiently, avoiding common performance bottlenecks. By integrating these nodes into your ComfyUI setup, you can achieve more flexible and efficient image processing workflows.

zhangp365/ComfyUI-utils-nodes Features

LoadImageWithSwitch

This node is a modified version of the official LoadImage node. It includes a switch that, when turned off, prevents the image from being loaded. This can be useful when you want to conditionally load images based on certain criteria.

LoadImageMaskWithSwitch

Similar to LoadImageWithSwitch, this node adds a switch to the official LoadImageMask node. When the switch is off, the image mask is not loaded, allowing for more control over when masks are applied.

LoadImageWithoutListDir

This node improves performance when loading images from directories with a large number of files. It avoids using os.listdir, which can be slow, making the image loading process faster.

LoadImageMaskWithoutListDir

An optimized version of the LoadImageMask node that avoids using os.listdir for better performance when dealing with large directories.

ImageCompositeMaskedWithSwitch

This node adds a switch to the official ImageCompositeMasked node. When the switch is off, the node returns the destination image directly, bypassing the compositing process.

ImageBatchOneOrMore

This node allows you to input one or more images, with a limit of six. It extends the functionality of the official ImageBatch node, which is limited to two images, making it easier to work with multiple images simultaneously.

ImageConcatenateOfUtils

An extension of the original ImageConcatenate node, this node adds the capability to upscale images and includes additional checks to determine if the second image is empty.

ColorCorrectOfUtils

An extension of the original ColorCorrect node, this node adds functionalities for adjusting the red, green, and blue channels.

ModifyTextGender

This node adjusts text to describe gender based on the input. If the input is 'M', the text is adjusted to describe a male; if 'F', it describes a female.

GenderControlOutput

This node determines the output based on the input gender. It outputs gender-specific text, float, and integer values.

SplitMask

This node splits one mask into two masks of the same size based on the area of the submasks. If there are more than two areas, it selects the two largest submasks.

MaskFastGrow

Designed for quickly growing masks, this node maintains consistent speed regardless of the grow value, unlike other mask growth nodes that slow down with large grow values.

MaskFromFaceModel

Generates a mask from the face model of the Reactor face swap node output. The mask covers the facial area below the eyes, excluding the forehead.

MaskAutoSelector

Checks three input masks and returns the first available one. If none are available, it raises an exception.

MaskCoverFourCorners

Generates a mask by covering the selected corners with circular edges, useful for removing watermarks from the corners.

CheckpointLoaderSimpleWithSwitch

Enhances the official LoadCheckpoint node by integrating three switches to control whether specific components are loaded, saving memory when certain components are not needed.

ImageResizeTo8x

A modified version of the image-resize-comfyui node, this node adds logic to crop the resulting image size to 8 times size, similar to the VAE encode node, to avoid pixel differences when pasting back by the ImageCompositeMasked node.

TextPreview

A convenient node for previewing text, originally from ComfyUI-Custom-Scripts.

MatchImageRatioToPreset

Determines the closest standard SDXL training size based on the input image ratio, useful for subsequent image resizing and other processes.

UpscaleImageWithModelIfNeed

Enhances the official UpscaleImageWithModel node by adding a condition to bypass upscaling if the input image area exceeds a predefined threshold.

DetectorForNSFW

Adapts the original model and inference code from nudenet for use with Comfy. It includes a small 10MB default model and allows for adjusting confidence levels for various rules.

DeepfaceAnalyzeFaceAttributes

Integrates the deepface library to analyze face attributes such as gender, race, emotion, and age. It processes one image at a time and supports a standard square face image mode.

Troubleshooting zhangp365/ComfyUI-utils-nodes

Common Issues and Solutions

Issue: Node Not Loading

• Solution: Ensure that the node is correctly installed in the custom_nodes folder and that all dependencies are installed. Restart ComfyUI after installation.

Issue: Slow Performance

• Solution: Use nodes like LoadImageWithoutListDir and LoadImageMaskWithoutListDir to improve performance when dealing with large directories.

Issue: Incorrect Mask Generation

• Solution: Verify the input parameters and ensure that the correct mask node is being used. For specific mask requirements, use nodes like MaskFromFaceModel or MaskCoverFourCorners.

Frequently Asked Questions

Q: How do I enable or disable a switch in a node?

• A: Each node with a switch will have a toggle option in its settings. Simply turn the switch on or off as needed.

Q: Can I use multiple images with ImageBatchOneOrMore?

• A: Yes, you can input up to six images using this node.

Q: How do I adjust the confidence levels for NSFW detection?

• A: You can adjust the confidence levels in the node settings to be more lenient or strict based on your requirements.

Learn More about zhangp365/ComfyUI-utils-nodes

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

• ComfyUI Documentation
• ComfyUI Community Forums
• Deepface Library
• Image Resize for ComfyUI These resources provide valuable information and support to help you get the most out of ComfyUI-utils-nodes.