ComfyUI Node: 📈 RGB Curves

📈 RGB Curves:

The CurvesNode is a powerful tool designed to enhance and manipulate images by adjusting their color channels through customizable curve transformations. This node allows you to apply precise adjustments to the RGB channels or individual Red, Green, and Blue channels of an image, enabling you to achieve specific visual effects or correct color imbalances. By defining control points on a curve, you can dictate how input values are mapped to output values, offering a high degree of control over the image's tonal range. The node supports different interpolation methods, such as linear, cubic, and quadratic, to smoothly transition between control points, ensuring flexibility in achieving the desired effect. Additionally, the node provides options to adjust the strength of the effect and preserve the original image's luminosity, making it a versatile tool for both creative and corrective image processing tasks.

📈 RGB Curves Input Parameters:

image

The image parameter is the input image that you want to process. It serves as the base upon which the curve adjustments will be applied. This parameter is essential as it provides the data that will be transformed by the node.

channel

The channel parameter specifies which color channels of the image will be affected by the curve adjustments. You can choose from "RGB" to apply changes to all color channels, or select "Red", "Green", or "Blue" to target a specific channel. The default value is "RGB", allowing for comprehensive color adjustments.

curve_points

The curve_points parameter allows you to define the control points for the curve transformation. These points are specified as a string in the format x,y;x,y;..., where each pair represents a point on the curve. The default value is "0,0;64,64;128,128;192,192;255,255", which represents a linear transformation. This parameter is crucial for customizing the tonal mapping of the image.

interpolation

The interpolation parameter determines the method used to interpolate between the control points on the curve. You can choose from "linear", "cubic", or "quadratic" interpolation, with "cubic" being the default. This choice affects the smoothness and nature of the transitions between points, impacting the overall effect on the image.

strength

The strength parameter controls the intensity of the curve effect applied to the image. It is a float value ranging from 0.0 to 2.0, with a default of 1.0. A value of 0.0 means no effect, while 2.0 doubles the effect's intensity. This parameter allows you to fine-tune the impact of the curve adjustments.

preserve_luminosity

The preserve_luminosity parameter is a boolean option that, when enabled, maintains the original luminosity of the image while applying the curve adjustments. The default value is False. This option is useful for ensuring that the overall brightness of the image remains consistent, even after color transformations.

📈 RGB Curves Output Parameters:

IMAGE

The output of the CurvesNode is an IMAGE that has been processed according to the specified curve adjustments. This output reflects the changes made to the color channels based on the input parameters, providing a visually enhanced or corrected version of the original image. The output image retains the same dimensions and format as the input, ensuring compatibility with subsequent processing steps.

📈 RGB Curves Usage Tips:

• To achieve subtle color corrections, start with a gentle curve and gradually increase the strength parameter to see the effect incrementally.
• Use the preserve_luminosity option when you want to maintain the original brightness of the image, especially when making significant color adjustments.
• Experiment with different interpolation methods to find the one that best suits the desired effect, as each method can produce distinct visual results.

📈 RGB Curves Common Errors and Solutions:

Invalid curve_points format

• Explanation: The curve_points string is not formatted correctly, which can occur if the points are not separated by semicolons or if the coordinates are not separated by commas.
• Solution: Ensure that the curve_points string follows the correct format, such as x,y;x,y;..., and that all values are valid numbers.

Image tensor shape mismatch

• Explanation: The input image does not have the expected shape, which can happen if the image is not properly formatted as a tensor.
• Solution: Verify that the input image is a correctly formatted tensor with the appropriate dimensions, typically a 3D array for single images or a 4D array for batches.

Interpolation method not supported

• Explanation: An unsupported interpolation method is specified, which can occur if the method name is misspelled or not among the available options.
• Solution: Check that the interpolation parameter is set to one of the supported methods: "linear", "cubic", or "quadratic".