ComfyUI | Alessandro Zonta

Enhancing Image Cropping with ComfyUI-CenterNode

Tue, 11 Jun 2024 00:00:00 +0000

Enhancing Image Cropping with ComfyUI-CenterNode

The ComfyUI-CenterNode, developed by Alessandro Zonta, introduces the “Bounding Box Crop” node for precise image cropping within the ComfyUI framework. This node computes the top-left coordinates of a cropped bounding box based on user-defined input coordinates and dimensions, ensuring accurate and centered cropping.

How It Works

Input Coordinates: Users provide the left, top, right, and bottom coordinates of the bounding box, along with the desired width and height of the cropped area.
Center Calculation: The node calculates the center of the cropping area, then determines the top-left coordinates for the final cropped image.
Output Coordinates: Outputs the computed coordinates for further use in image processing workflows.

The Idea Behind It

The Bounding Box Crop node aims to simplify and enhance the process of cropping images in ComfyUI. By ensuring precise calculation of cropping coordinates, it aids in creating centered and accurately cropped images, which is crucial for various applications such as object detection, image analysis, and graphic design.

Installation

To integrate the Bounding Box Crop node into ComfyUI, follow these steps:

Clone the repository into your ComfyUI custom_nodes directory:
```
git clone https://github.com/alessandrozonta/ComfyUI-CenterNode.git
```
or find the Custom Node from your ComfyUI-Manager

Conclusion

The ComfyUI-CenterNode is an essential tool for anyone needing precise and centered image cropping. By accurately computing the necessary coordinates, it enhances the efficiency and accuracy of image processing tasks, making it a valuable addition to the ComfyUI toolkit.

For more details, visit the GitHub page.

Introducing the LayersSaver Node for ComfyUI

Tue, 11 Jun 2024 00:00:00 +0000

Introducing the LayersSaver Node for ComfyUI

The LayersSaver nodes, developed by Alessandro Zonta, is a powerful custom addition to ComfyUI, designed to enhance image manipulation workflows. Two nodes are available on the packadge. The first node allows users to create and manage image layers based on input masks, saving the results into a PSD file. The second node does a similar thing but only from th einput images. It can receive a batched image and every image in the batch is going to be saved as a separate layer on the ouput PSD. This functionality is particularly valuable for graphic design, image processing, and other tasks requiring layer-based image manipulation.

How It Works

Input Image Handling: The first node processes a single input image, ensuring the removal of the batch dimension for simplicity. The second nodes processes a infinite bach dimension input image.
Mask Application: Multiple masks are applied to the input image, creating distinct layers for each masked region.
Original Image Inclusion: Users can optionally include the original image as a separate layer in the PSD file.
Alpha Channel Management: Ensures the input image includes an alpha channel, facilitating transparency.
Dynamic PSD Naming: If a filename isn’t specified, the node generates a unique PSD filename based on the current timestamp.
PSD File Creation: The final image layers are saved into a PSD file, which can then be used in various image editing software.

The Idea Behind It

The idea behind the LayersSaver node is to streamline the process of creating and managing image layers within ComfyUI. By allowing users to segment images into separate layers based on masks, it provides a high degree of flexibility and control over image editing. This is especially useful in professional graphic design and advanced image processing tasks, where layer manipulation is crucial.

Installation

To integrate the LayersSaver node into ComfyUI, follow these steps:

Clone the repository into your ComfyUI custom_nodes directory:
```
git clone https://github.com/alessandrozonta/ComfyUI-Layers
```
or find the Custom Node from your ComfyUI-Manager

Install the necessary libraries by running:

pip install pytoshop -I --no-cache-dir
pip install psd-tools --no-deps

Restart ComfyUI to load the new node.

Conclusion

The LayersSaver node is an invaluable tool for anyone involved in graphic design or image processing. By providing a streamlined method for creating and managing image layers, it enhances the functionality of ComfyUI, making complex image editing tasks more efficient and manageable. Whether you’re a professional designer or a hobbyist, this node offers the flexibility and power needed to take your projects to the next level.

For more details, visit the GitHub page.

Unleashing the Power of Human Pose Estimation with ComfyUI-OpenPose

Tue, 11 Jun 2024 00:00:00 +0000

Unleashing the Power of Human Pose Estimation with ComfyUI-OpenPose

The ComfyUI-OpenPose node, created by Alessandro Zonta, brings advanced human pose estimation capabilities to the ComfyUI ecosystem. This custom node leverages OpenPose models to extract and visualize human pose keypoints from input images, enhancing image processing and analysis workflows.

How It Works

Model Integration: The node downloads OpenPose models from the Hugging Face Hub and saves them locally.
Image Processing: Processes a single input image to extract human pose keypoints, ensuring batch processing is not required.
Visualization: Draws keypoints and limbs on the original image with adjustable transparency, providing clear visual representation.
Alternative Outputs: Generates an image with keypoints on a black background and exports keypoints data in JSON format.
Model Options: Users can choose between two OpenPose representations: COCO and BODY_25.

The Idea Behind It

The primary goal of the ComfyUI-OpenPose node is to simplify and enhance human pose estimation within the ComfyUI framework. By integrating OpenPose models, this node allows for precise keypoint extraction and visualization, making it invaluable for applications in computer vision, animation, sports analysis, and more.

Installation

To integrate the OpenPose node into ComfyUI, follow these steps:

Clone the repository into your ComfyUI custom_nodes directory:
```
git clone https://github.com/alessandrozonta/ComfyUI-OpenPose.git
```
or find the Custom Node from your ComfyUI-Manager

Install the necessary libraries by running:

pip install -r requirements.txt

python install.py

Restart ComfyUI to load the new node.

Conclusion

The ComfyUI-OpenPose node is a powerful addition for anyone looking to incorporate human pose estimation into their image processing tasks. By providing robust keypoint extraction and visualization capabilities, it significantly enhances the functionality of ComfyUI, making it a valuable tool for both professionals and enthusiasts in various fields.

For more details, visit the GitHub page.