ComfyUI Extension: ComfyUI-ResizeZeptaPayload

Authored by Pablerdo

Created

Updated

1 stars

Resize a batch of trajectories or images

Custom Nodes (0)

    README

    ComfyUI-ResizeZeptaPayload

    A custom node extension for ComfyUI that provides utilities for resizing images and trajectory data while maintaining compatibility with Zepta AI tools.

    Features

    • ResizeImageBatch: Resize a batch of images with various methods and settings

      • Control output dimensions with multiple input options
      • Keep proportions while resizing
      • Make dimensions divisible by a specific value
      • Choose from multiple upscaling methods (nearest-exact, bilinear, area, bicubic, lanczos)
      • Center cropping option
    • ResizeTrajectoryBatch: Resize trajectory coordinates proportionally to match resized images

      • Automatically scales trajectory points to new dimensions
      • Maintains the relative positioning of trajectory data

    Installation

    1. Clone this repository into your ComfyUI custom_nodes directory:

      cd path/to/ComfyUI/custom_nodes
      git clone https://github.com/yourusername/ComfyUI-ResizeZeptaPayload.git
      
    2. Restart ComfyUI

    Usage

    ResizeImageBatch

    The node can be found in the node menu under the "ResizeZeptaPayload" category.

    Input parameters:

    • images: Batch of images to resize
    • output_width: Target width (set to 0 to maintain original width)
    • output_height: Target height (set to 0 to maintain original height)
    • upscale_method: Interpolation method to use
    • keep_proportion: Whether to maintain aspect ratio
    • divisible_by: Make dimensions divisible by this value
    • crop: Crop method (disabled or center)
    • width_input (optional): Override output width
    • height_input (optional): Override output height
    • get_size_from_image (optional): Get dimensions from this image

    ResizeTrajectoryBatch

    Input parameters:

    • input_width: Original width of the context
    • input_height: Original height of the context
    • output_width: Target width
    • output_height: Target height
    • trajectories: JSON string containing trajectory data