ComfyUI Extension: ComfyUI Equirectangular Tools

Authored by kukuo6666

Created

Updated

1 stars

Tools for processing equirectangular images, supporting conversion from equirectangular format to cubemap.

Custom Nodes (0)

    README

    ComfyUI-Equirect

    ComfyUI nodes for conversion between equirectangular panoramas and cubemaps.

    Features

    • EquirectToCubemapNode: Convert equirectangular panoramas to 6 cubemap faces
    • CubemapToEquirectNode: Convert 6 cubemap faces back to equirectangular panoramas

    Installation

    1. Clone this repository to your ComfyUI custom_nodes directory:

      cd ComfyUI/custom_nodes
      git clone https://github.com/yourusername/ComfyUI-Equirect.git
      
    2. Install dependencies:

      cd ComfyUI-Equirect
      pip install -r requirements.txt
      

    Usage

    EquirectToCubemapNode (Equirectangular to Cubemap)

    Inputs:

    • equirect_image: Equirectangular panorama input (2:1 aspect ratio)
    • face_size: Size of the output cubemap faces
    • fov: Field of view angle (default: 90 degrees)

    Outputs:

    • front, right, back, left, top, bottom: 6 cubemap faces

    CubemapToEquirectNode (Cubemap to Equirectangular)

    Inputs:

    • front, right, back, left, top, bottom: 6 cubemap faces
    • output_height: Height of the output equirectangular image (width will be automatically set to twice the height)

    Outputs:

    • equirect_image: Converted equirectangular panorama

    Technical Notes

    • Uses py360convert library for high-quality conversion when available
    • Automatically falls back to custom implementation if the library is not available
    • Supports batch processing and maintains correct image formats

    Dependencies

    • torch
    • numpy
    • pillow
    • opencv-python
    • py360convert

    Parameters

    • Input Image: Equirectangular panorama image with 2:1 aspect ratio
    • Face Size: Edge length for each cubemap face (default: 512, range: 64-4096)
    • Field of View: FOV angle for each face (default: 90, range: 60-120)

    System Requirements

    • ComfyUI
    • Python 3.7+
    • PyTorch
    • Pillow (PIL)
    • NumPy

    License

    This project is licensed under the MIT License - see the LICENSE file for details