This repository provides a collection of custom nodes for ComfyUI that leverage the functionalities of the pytorch360convert library, enabling efficient manipulation of 360-degree panoramic images. Users can convert between various panoramic formats, enhancing their workflow when working with equirectangular images and cubemaps.
- Allows conversion between equirectangular images and cubemaps, facilitating different viewing experiences.
- Includes nodes for cropping and pasting image sections, optimizing memory use during inpainting tasks.
- Features advanced functionalities like seamless image padding and face extraction for improved image quality.
Context
This tool is designed to extend the capabilities of ComfyUI by integrating custom nodes that utilize the pytorch360convert library. Its primary purpose is to provide users with advanced tools for working with 360-degree images, making it easier to convert between formats and manipulate panoramic imagery.
Key Features & Benefits
The custom nodes introduced in this repository include essential functions for converting equirectangular images to cubemaps and vice versa. Additionally, users can rotate equirectangular images along multiple axes, crop images for focused edits, and create masks for inpainting, all of which streamline the image editing process in ComfyUI.
Advanced Functionalities
This toolkit offers advanced capabilities such as the ability to apply circular padding to reduce seams in images, as well as specialized nodes for extracting and reconstructing faces from equirectangular images. These features are particularly useful for projects requiring high-quality image transformations and seamless stitching.
Practical Benefits
By integrating these custom nodes into their workflows, users can achieve greater control over image transformations, leading to improved quality and efficiency in their projects. The ability to manipulate images in a variety of ways reduces the time and effort needed to achieve desired results in 360-degree imaging.
Credits/Acknowledgments
This project is developed by Ben Egan, and it is licensed under the MIT License. Contributions to the repository are welcomed, and users are encouraged to cite the work appropriately in their research or projects.