ComfyUI Node: hiresfixScale

hiresfixScale:

The ttN hiresfixScale node is designed to upscale images using advanced methods and models, ensuring high-quality results while maintaining the original aspect ratio. This node is particularly useful for AI artists who need to enhance the resolution of their images without compromising on quality. It offers various rescaling options, including percentage-based scaling and aspect ratio maintenance, making it versatile for different artistic needs. The node also supports post-model rescaling and can encode the output into a latent space if required. By leveraging sophisticated upscaling techniques, ttN hiresfixScale helps you achieve detailed and crisp images, perfect for high-resolution outputs.

hiresfixScale Input Parameters:

upscale_model

Specifies the model to be used for upscaling the image. This parameter is crucial as it determines the quality and characteristics of the upscaled image. The available options are listed in the upscale_models directory, and the default value is None.

image

The input image that you want to upscale. This parameter accepts an image file and is the primary input for the upscaling process.

rescale_after_model

A boolean parameter that determines whether to rescale the image after the model upscaling. If set to True, the image will be rescaled based on the specified method and dimensions. The default value is True.

rescale_method

Defines the method to be used for rescaling the image. The available methods include nearest-exact, bilinear, area, bicubic, lanczos, and bislerp. This parameter affects the smoothness and quality of the rescaled image.

rescale

Specifies the rescaling approach. Options include by percentage, to Width/Height, and `to longer side

• maintain aspect`. This parameter determines how the image dimensions will be adjusted during rescaling.

percent

An integer parameter that sets the percentage by which to scale the image. It is used when rescale is set to by percentage. The value ranges from 0 to 1000, with a default of 50.

width

Specifies the target width for the image when rescaling. This parameter is used when rescale is set to to Width/Height. The value ranges from 64 to the maximum resolution defined by MAX_RESOLUTION, with a default of 1024.

height

Specifies the target height for the image when rescaling. This parameter is used when rescale is set to to Width/Height. The value ranges from 64 to the maximum resolution defined by MAX_RESOLUTION, with a default of 1024.

longer_side

Defines the length of the longer side of the image when rescaling while maintaining the aspect ratio. This parameter is used when rescale is set to `to longer side

• maintain aspect. The value ranges from 64 to the maximum resolution defined by MAX_RESOLUTION`, with a default of 1024.

crop

Determines whether to crop the image during rescaling. The available options are disabled and center. This parameter affects how the image is adjusted to fit the target dimensions.

image_output

Specifies how the output image should be handled. Options include Hide, Preview, Save, and Hide/Save. This parameter controls the visibility and storage of the upscaled image.

save_prefix

A string parameter that sets the prefix for the saved image file. The default value is ComfyUI.

output_latent

A boolean parameter that determines whether to encode the output image into a latent space. If set to True, the image will be encoded, which can be useful for further processing. The default value is True.

hiresfixScale Output Parameters:

latent

The encoded latent representation of the upscaled image. This output is available if output_latent is set to True. It can be used for further processing or analysis in latent space.

image

The upscaled image. This output provides the final high-resolution image after applying the specified upscaling and rescaling methods. It is the primary output for visual inspection and use.

hiresfixScale Usage Tips:

• For best results, choose an upscale model that matches the style and content of your input image.
• Use the rescale_after_model option to fine-tune the dimensions of the upscaled image, ensuring it fits your desired resolution.
• Experiment with different rescale_method options to find the best balance between quality and performance for your specific use case.
• When maintaining the aspect ratio, use the longer_side parameter to ensure the image scales proportionally without distortion.

hiresfixScale Common Errors and Solutions:

Unable to model upscale. Please install an upscale model and try again.

• Explanation: This error occurs when no upscale model is specified or the specified model is not installed.
• Solution: Ensure that you have selected a valid upscale model from the upscale_models directory and that it is properly installed.

Image dimensions exceed maximum resolution.

• Explanation: This error occurs when the specified width or height exceeds the maximum allowed resolution.
• Solution: Adjust the width, height, or longer_side parameters to values within the allowed range defined by MAX_RESOLUTION.

Invalid rescale percentage.

• Explanation: This error occurs when the percent parameter is set to a value outside the allowed range.
• Solution: Ensure that the percent parameter is set to a value between 0 and 1000.