Visit ComfyUI Online for ready-to-use ComfyUI environment
Enhance and modify image areas seamlessly with advanced inpainting techniques for precise control and natural blending.
Fooocus Inpaint is a powerful node designed to enhance and modify specific areas of an image by intelligently filling in or altering the selected regions. This node is particularly useful for AI artists who want to refine their artwork by removing unwanted elements, repairing damaged areas, or adding new details seamlessly. By leveraging advanced inpainting techniques, Fooocus Inpaint ensures that the modifications blend naturally with the surrounding pixels, maintaining the overall aesthetic and coherence of the image. The node offers a range of customizable parameters, allowing you to control the inpainting process with precision and achieve the desired results effectively.
This parameter represents the input image that you want to inpaint. It is the primary canvas on which the inpainting operations will be performed. The image should be provided in a compatible format that the node can process.
This boolean parameter determines whether the initial latent space should be disabled during the inpainting process. When set to True
, the initial latent space is not used, which can affect the quality and style of the inpainting. The default value is False
.
This float parameter controls the respective field for the inpainting process. It influences how the inpainting algorithm considers the surrounding pixels to fill in the selected area. The value ranges from 0 to 1.0, with a default of 0.618. Adjusting this parameter can help achieve more natural and coherent inpainting results.
This parameter specifies the version of the inpainting engine to be used. Different versions may offer varying levels of performance and quality. The default value is "v2.6"
, and you can choose from the available versions provided by the modules.flags.inpaint_engine_versions
.
This boolean parameter indicates whether the top part of the image should be outpainted. When set to True
, the node will extend the image upwards, filling in the new area based on the existing content. The default value is False
.
This boolean parameter indicates whether the bottom part of the image should be outpainted. When set to True
, the node will extend the image downwards, filling in the new area based on the existing content. The default value is False
.
This boolean parameter indicates whether the left part of the image should be outpainted. When set to True
, the node will extend the image to the left, filling in the new area based on the existing content. The default value is False
.
This boolean parameter indicates whether the right part of the image should be outpainted. When set to True
, the node will extend the image to the right, filling in the new area based on the existing content. The default value is False
.
This optional parameter represents the mask that defines the area of the image to be inpainted. The mask should be provided in a compatible format, where the regions to be inpainted are marked. If no mask is provided, the node will attempt to determine the inpainting area automatically.
This output parameter returns the inpainted image along with the applied mask and other relevant settings. The output is a dictionary containing the modified image, the mask used, and the configuration parameters that were applied during the inpainting process. This allows you to review and further process the inpainted image as needed.
inpaint_respective_field
parameter to find the optimal setting for your image. A value closer to 1.0 considers more surrounding pixels, which can result in smoother and more natural inpainting.inpaint_disable_initial_latent
parameter to control the influence of the initial latent space. Disabling it can sometimes lead to more creative and unexpected results, but may also affect the coherence of the inpainting.modules.flags.inpaint_engine_versions
and select a supported version.© Copyright 2024 RunComfy. All Rights Reserved.