ComfyUI Node: Blur images

Blur images:

The LF_BlurImages node is designed to apply a Gaussian blur effect to a set of input images, providing a smooth and aesthetically pleasing transformation. This node is particularly useful for AI artists looking to soften the details of their images, create a dreamy or out-of-focus effect, or prepare images for further processing where reduced detail is beneficial. By adjusting the blur intensity based on the image size, the node ensures that the blur effect is proportionate and visually consistent across different images. The node processes each image individually, applies the blur, and then returns the blurred images along with their new filenames, making it easy to manage and utilize the transformed images in subsequent steps of your workflow.

Blur images Input Parameters:

images

This parameter expects a list of image tensors that you want to apply the blur effect to. Each image tensor should be in the format [1, H, W, 3], where H is the height, W is the width, and 3 represents the RGB channels. The images are processed individually, and the blur effect is applied based on the specified blur percentage.

file_names

This parameter is a list of filenames corresponding to the input images. These filenames are used to generate new filenames for the blurred images by appending a _Blur suffix to the original names. This helps in easily identifying and managing the blurred images.

blur_percentage

This parameter determines the intensity of the blur effect. It can be a single value or a list/tuple of values, but only the first value will be used if multiple values are provided. The blur percentage is scaled based on the minimum dimension of each image to ensure a consistent visual effect. The higher the blur percentage, the more pronounced the blur effect will be.

Blur images Output Parameters:

blurred_images

This output is a list of image tensors that have been processed and blurred. Each tensor corresponds to an input image, but with the Gaussian blur effect applied. These tensors can be used in further processing steps or saved as new image files.

blurred_file_names

This output is a list of filenames for the blurred images. Each filename is derived from the corresponding input filename by appending a _Blur suffix. This makes it easy to distinguish between the original and blurred images.

Blur images Usage Tips:

• To achieve a subtle blur effect, use a lower blur percentage. This is useful for slightly softening the details without losing the overall structure of the image.
• For a more dramatic, dreamy effect, increase the blur percentage. This can be particularly effective for background elements or to create a sense of depth.
• Ensure that the input images are of high quality and resolution, as the blur effect can sometimes highlight imperfections in lower-quality images.

Blur images Common Errors and Solutions:

"Invalid image tensor format"

• Explanation: This error occurs if the input image tensors are not in the expected format [1, H, W, 3].
• Solution: Ensure that your input images are correctly formatted as tensors with the shape [1, H, W, 3].

"File names list length mismatch"

• Explanation: This error happens when the number of filenames provided does not match the number of input images.
• Solution: Verify that the length of the file_names list matches the length of the images list.

"Blur percentage out of range"

• Explanation: This error is raised if the blur percentage is not within a reasonable range.
• Solution: Ensure that the blur percentage is a positive value and within a practical range for your specific use case.