ComfyUI Node: Load Images & Compose (obvpm)

LoadImagesCompose (obvpm):

The LoadImagesCompose node is designed to streamline the process of combining multiple images into a single, cohesive reference frame. This node is particularly useful for AI artists who need to create a composite image from several individual pictures, such as characters, locations, and props, to be used as a single input for models that require one image. By integrating the functionalities of loading, cropping, and stitching images, this node simplifies the workflow, eliminating the need for multiple separate nodes. It automatically arranges the images in a layout that maintains their original order and aspect ratios, ensuring that the final composite image is tightly packed without unnecessary enlargement or rotation of the individual images. This makes it an efficient tool for creating reference sheets that fit within a specified megapixel budget, providing a seamless and user-friendly experience for artists.

LoadImagesCompose (obvpm) Input Parameters:

layers

The layers parameter is a JSON object that defines the images to be loaded and their respective cropping details. Each entry in the JSON array specifies an image file and an optional crop, which is defined by normalized coordinates. If the crop is absent or null, the entire image is used. This parameter allows you to manage multiple images and their compositions in a structured manner, ensuring that each image is placed correctly within the final composite. The JSON format provides flexibility and precision in defining how each image should be handled.

max_megapixels

The max_megapixels parameter sets the maximum size for the composite image in terms of megapixels. This ensures that the final output does not exceed a specified resolution, which is crucial for managing file sizes and processing power. The value must be a finite number between 0 and 128, with 0 indicating no limit. This parameter helps maintain the quality and manageability of the output image, especially when working with high-resolution inputs.

target_aspect

The target_aspect parameter defines the desired aspect ratio for the final composite image. It can be a named aspect or a finite ratio, ensuring that the output image fits a specific format or display requirement. This parameter is essential for creating images that need to conform to particular dimensions, such as those required for specific print or digital formats.

gap

The gap parameter specifies the amount of space, in pixels, between the images in the composite. It is an integer value ranging from 0 to 256, allowing you to control the spacing between images for aesthetic or practical purposes. This parameter is useful for creating visually appealing layouts or ensuring that images do not overlap in the final composition.

LoadImagesCompose (obvpm) Output Parameters:

image

The image output parameter represents the final composite image that contains all the input layers arranged according to the specified parameters. This single image output is the result of loading, cropping, and composing the input images into a cohesive reference frame. It is ready for use in applications that require a single image input, providing a convenient and efficient way to handle multiple images in one go.

LoadImagesCompose (obvpm) Usage Tips:

• Use the layers parameter to precisely define the images and their crops, ensuring that each image is placed correctly in the final composite.
• Adjust the max_megapixels parameter to manage the resolution and file size of the output image, especially when working with high-resolution inputs.

LoadImagesCompose (obvpm) Common Errors and Solutions:

Load Images & Compose: no images. Add one with the + button on the node, or drop image files onto it.

• Explanation: This error occurs when no images have been added to the node for composition.
• Solution: Ensure that you add at least one image to the node using the + button or by dragging and dropping image files onto the node.

target_aspect must be a named aspect or finite ratio

• Explanation: This error indicates that the target_aspect parameter has been set to an invalid value.
• Solution: Verify that the target_aspect is either a recognized named aspect or a valid finite ratio, and adjust the parameter accordingly.