Batch Uncrop Advanced (BatchUncropAdvanced) - ComfyUI-KJNodes

The Batch Uncrop Advanced node receives the original image, the cropped image, the corresponding crop mask, a combined mask, and bounding box data as inputs.

The Batch Uncrop Advanced node is an advanced image processing node specifically designed to accurately re-integrate cropped and modified image fragments back into the original background image. Its core functionality is to process images in batches and supports the use of masks and bounding boxes to control the blending process, achieving seamless compositing results.

Node Functionality

This node receives the original image, the cropped image, the corresponding crop mask, a combined mask, and bounding box data as inputs. Its technical implementation relies on these masks and bounding box information to locate and blend the images. It supports feathering the blend edges (via the border_blending parameter) and allows scaling of the cropped area (via the crop_rescale parameter) to fine-tune the compositing result.

Node Parameter Description - Batch Uncrop Advanced

Connection Inputs (Inputs)

Parameter Name	Data Type	Required	Default Value	Range/Options	Description
`original_images`	IMAGE	Yes	-	-	original_images (Input Parameter): The original full image serving as the background, onto which the cropped images will be blended. The number of input images must match the number of `cropped_images`.
`cropped_images`	IMAGE	Yes	-	-	cropped_images (Input Parameter): The cropped image fragments that need to be pasted back onto the original background. The quantity must be the same as `original_images`.
`cropped_masks`	MASK	Yes	-	-	cropped_masks (Input Parameter): The mask corresponding to each `cropped_image`, used to define the valid area within each cropped fragment that needs to be blended.
`combined_crop_mask`	MASK	Yes	-	-	combined_crop_mask (Input Parameter): A combined mask, typically formed by merging the masks of all cropped areas, used to define the overall pasteable region.
`bboxes`	BBOX	Yes	-	-	bboxes (Input Parameter): A list of bounding boxes specifying the precise location and extent within the `original_images` where each `cropped_image` should be placed.
`combined_bounding_box`	BBOX	No	-	-	combined_bounding_box (Input Parameter): An optional bounding box that defines the overall region for the combined mask. If not provided, the bounding box is derived from the individual `bboxes`.

Control Parameters (Parameters)

Parameter Name	Data Type	Required	Default Value	Range/Options	Description
`border_blending`	FLOAT	Yes	0.25	0.0 - 1.0 (Step: 0.01)	border_blending (Input Parameter): Controls the feathering intensity of the blend boundary, range 0.0 to 1.0, step 0.01. A higher value results in a smoother edge transition.
`crop_rescale`	FLOAT	Yes	1.0	0.0 - 10.0 (Step: 0.01)	crop_rescale (Input Parameter): The scaling factor applied to `cropped_images`, range 0.0 to 10.0, step 0.01. 1.0 means maintaining the original size.
`use_combined_mask`	BOOLEAN	Yes	False	-	use_combined_mask (Input Parameter): Boolean value. When True, uses `combined_crop_mask` for global blending control; when False, uses individual `cropped_masks`.
`use_square_mask`	BOOLEAN	Yes	True	-	use_square_mask (Input Parameter): Boolean value. When True, processes the mask shape into a square; when False, retains the original mask shape.

Output

Parameter Name	Data Type	Description
IMAGE	IMAGE	IMAGE (Output Parameter): Outputs the final composited full image, with all cropped fragments blended into the original background.

Usage Scenarios

A typical application is within "inpainting" or "image restoration" workflows. For example, a user can first use other nodes (such as a Bounding Box node) to select multiple areas from the original image that require modification. After redrawing or editing these areas separately, this node can be used to paste all modified areas back into the original image at once and naturally, thereby efficiently completing multi-target editing.

Notes

When using this node, please note that the number of images in the input original_images and cropped_images groups must be strictly identical. If more bounding boxes (bboxes) are provided than there are images, the extra bounding boxes will be automatically ignored.

Batch Uncrop Advanced Node Source Code Link

The Batch Uncrop Advanced node is from the ComfyUI-KJNodes node package.

Batch Uncrop Advanced(BatchUncropAdvanced)