A straightforward tool for ComfyUI, this repository provides nodes that enable users to resize images according to specified pixel counts and maintain consistent resolutions. It focuses on adjusting the overall pixel count, side lengths, and size ratios rather than just the width and height.
- Allows resizing of images to a defined pixel resolution.
- Maintains or adjusts the image's aspect ratio as required.
- Ensures that dimensions align with the greatest common divisor (GCD), typically 32 or 64.
Context
This tool serves as a set of nodes within ComfyUI designed specifically for image resizing tasks. Its primary function is to facilitate the scaling of images to a desired total pixel count while ensuring that the output dimensions adhere to specific standards.
Key Features & Benefits
The resizing nodes in this repository provide practical functionality by allowing users to specify pixel resolutions directly. Users can either maintain the original aspect ratio of images or adjust to a new one based on their requirements, which is crucial for ensuring that images retain their visual integrity after resizing. Additionally, the nodes guarantee that the dimensions of resized images are multiples of the GCD, which is important for compatibility in various applications.
Advanced Functionalities
This tool includes advanced capabilities such as the ability to crop images and resize them by ratio, which offers users flexibility in how they manipulate their images. These functionalities are particularly useful for users who require precise control over image dimensions for specific projects or workflows.
Practical Benefits
By integrating these nodes into their workflow, users can enhance their efficiency in image processing within ComfyUI. The tool simplifies the task of resizing images, allowing for more streamlined operations and better control over the final output quality. This results in a more effective use of time and resources when working with images in AI art projects.
Credits/Acknowledgments
The project is developed by MakkiShizu and is released under the MIT License, allowing for open use and modification.