cg-image-filter

Any List to String

Compact description unavailable for this node.

Batch from Image List

Compact description unavailable for this node.

1 input · 0 parameters

cg_String to Float

Compact description unavailable for this node.

0 inputs · 2 parameters

cg_String to Int

Compact description unavailable for this node.

0 inputs · 2 parameters

Image Filter

The image filter node pauses execution of the workflow while you choose which, if any, of the images produced, you want to progress. Insert it like this: When you run the workflow, and get to this point, a popup window will appear displaying the image(s) for you to select which, if any, you want to progress: Click the images that you want to keep (their border will turn green) and then click 'Send' to continue the workflow. If you don't want to keep any of the images, click 'Cancel' (or press 'escape') to terminate the workflow. The node also has a timeout specified, and a countdown is shown on the left hand side. If you don't Send or Cancel before the timeout, the node will either cancel or send all the images, depending on the option you have selected. Here's a simple use: generate a batch of images and pick which ones you want to save: If you hover over an image (it gets a red border) and press the space bar, it will zoom to fill the screen. You can also use the Enter the Image Filter node with an image zoomed setting to start in this mode. While zoomed, you can navigate with keyboard or mouse: | |Mouse|Keyboard|Notes| |-|-|-|-| |Select/unselect/send|Click the zoomed image|Arrow Up|Will select/unselect the image, or send it if you have Clicking an image sends it turned on| |Previous/Next|Click the arrows to the left or right|Arrow Left Arrow Right|The image number is shown near the top right| |Unzoom| |Space|Back to the grid view| The Latent and Mask inputs are optional. If used, they should have the same number of latents (or masks) as the image batch, and the latents…

3 inputs · 11 parameters

Image List From Batch

Compact description unavailable for this node.

1 input · 0 parameters

Mask Image Filter

Designed for a single image, when executed the Mask Image Filter node will automatically launch the mask editor. When you finish mask editing the image and mask will be output. Here's a simple use - generate an image, and then mask the bit you don't like, before doing an img2img step. Again, there is a timeout, and if you don't save a mask before the end of the timeout (or if you press the cancel button in the mask editor), it will either cancel, or send a blank mask, depending on the option chosen. There is an optional mask input, which allows you to specify the mask when the editor is launched: Note that the Mask Image Filter works with the new Mask Editor; it does not work with the old one There are four options for what to do if the inputs are exactly the same as before: - Run normally: use the input mask, if any, as the starting point - Start with last output: use the last mask sent as the starting point - Resend last output: don't launch the editor, just resend the output - Always start with last output: use the last mask sent even if the inputs have changed, as long as the image dimensions are the same

2 inputs · 9 parameters

Masked Section

Masked Section takes a mask and a (batch of) images and outputs the images cropped to the bounding box of the mask (with a minimum size). Here's how you might use it to preview the parts of the image that were changed in img2img (also using Pick from List and Image List from Batch described below)

2 inputs · 1 parameter

Pick from List

Compact description unavailable for this node.

Split String by Commas

Compact description unavailable for this node.

0 inputs · 2 parameters

StringToStringList

Splits an input string on the specified substring, or on line breaks when split is "linebreak"; trims whitespace from each part and outputs the parts as a string list.

0 inputs · 2 parameters

Text Image Filter

Also designed for a single image, this node will show the image and a string of text; you can edit the text and then press send. The image and (edited) text are output. One example use is for captioning workflows; you can read and edit each caption as it is generated. Here's a trivial workflow:

2 inputs · 9 parameters