ComfyUI Node: Corruptor | GlitchNodes

Corruptor | GlitchNodes:

The Corruptor node is designed to apply controlled corruption effects to images through the use of wavelet transformations. This node allows you to introduce artistic distortions to images by manipulating wavelet coefficients, which can be adjusted in intensity and applied across various color spaces. The primary goal of the Corruptor is to provide a flexible tool for creating glitch art effects, enabling you to explore creative possibilities by altering the visual structure of images. By leveraging wavelet transformations, the Corruptor decomposes an image, applies distortions, and reconstructs it, offering a unique method to achieve visually striking results. This node is particularly beneficial for AI artists looking to experiment with image aesthetics and introduce intentional imperfections or noise to their artwork.

Corruptor | GlitchNodes Input Parameters:

image

The input image to be corrupted. This parameter accepts a tensor representing the image data, which will be processed to apply the desired corruption effects.

scaling_factor_in

This parameter controls the scaling factor applied to the wavelet coefficients during the decomposition phase. A higher value results in more pronounced corruption effects. The minimum value is 0, and there is no explicit maximum, but typical values range from 1 to 100.

scaling_factor_out

This parameter determines the scaling factor applied to the wavelet coefficients during the reconstruction phase. It influences the intensity of the final corruption effect. Similar to scaling_factor_in, the minimum value is 0, and typical values range from 1 to 100.

noise_strength

This parameter specifies the strength of the noise added to the image. It affects the level of randomness introduced into the image, with higher values resulting in more noticeable noise. The minimum value is 0, and there is no explicit maximum, but typical values range from 0 to 100.

color_space

The color space in which the corruption is applied. Options include "HSV", "LAB", "YUV", and "RGB". This parameter determines how the image is processed and can significantly affect the visual outcome of the corruption.

channels_combined

A boolean parameter that indicates whether all color channels should be processed as a single data stream. If set to True, the channels are combined, leading to a more uniform corruption effect across the image. If False, each channel is processed independently.

wavelet_floor_mode

This parameter defines the method used for flooring wavelet coefficients. Options include "regular", "absolute", and "threshold". Each mode offers a different approach to handling the coefficients, affecting the symmetry and cleanliness of the corruption.

wavelet_padding

Specifies the padding method used during wavelet transformation. Options include "edge" and potentially others, which determine how the image edges are handled during processing.

wavelet_threshold

A threshold value applied to wavelet coefficients to control the level of detail preserved during corruption. Lower values retain more detail, while higher values result in more aggressive corruption.

noise_distribution

Determines the distribution type of the noise added to the image. Options include "normal", "uniform", and "salt_pepper", each providing a different noise pattern that can be used to achieve various artistic effects.

Corruptor | GlitchNodes Output Parameters:

corrupted_image

The output is a tensor representing the corrupted image. This image has undergone wavelet transformation, coefficient manipulation, and noise addition, resulting in a visually altered version of the input image. The corrupted image retains the original dimensions and color space, allowing for easy integration into further processing or display.

Corruptor | GlitchNodes Usage Tips:

• Experiment with different scaling_factor_in and scaling_factor_out values to achieve varying levels of distortion and artistic effects. Start with moderate values and adjust based on the desired intensity of the corruption.
• Utilize the color_space parameter to explore how different color representations affect the corruption outcome. Each color space can produce unique visual results, so try multiple options to find the most appealing effect for your artwork.
• Combine the noise_strength and noise_distribution parameters to introduce creative noise patterns. For subtle effects, use lower noise strengths with a "normal" distribution, while higher strengths and "salt_pepper" distribution can create more dramatic results.

Corruptor | GlitchNodes Common Errors and Solutions:

Error in image corruption: <error_message>

• Explanation: This error occurs when there is an issue during the image corruption process, possibly due to invalid input parameters or unexpected data types.
• Solution: Ensure that all input parameters are correctly specified and within their valid ranges. Check that the input image is in the correct format and dimensions expected by the node.

Error in corruption process: <error_message>

• Explanation: This error indicates a problem encountered during the corruption process, potentially due to incorrect parameter settings or processing logic.
• Solution: Review the parameter settings and ensure they are appropriate for the desired effect. Verify that the image data is compatible with the node's requirements and that all necessary dependencies are installed and functioning correctly.