This is the simple workflow I use to select elements in my videos and images. It's powerful when used correctly and can save a lot of time masking. This workflow can be combined with other workflows to achieve great results. I suggest expanding the mask a little bit before using the result (GrowMask and Mask Blur nodes in Comfy). There are many workflows here on CivitAI that use specific mask areas, which can be easily created using this workflow. It will produce three outputs: a video with the mask (black and white), a folder with the images of the mask (black and white), and a folder with the original video frames as images. This way, you can use the video directly in another workflow or load the image sequence in Comfy, A1111, or any other software you prefer to work with.
Description
-First Release
FAQ
Comments (6)
I'm glad you are back. Any plans to write an instruction for making the video you made 2months ago? I'm using "replacer" which does a similar thing, but the result is not as good as yours.
Yes. the workflows I'm releasing are part of what is needed to do it! Soon I might release thw whole workflow.
dope
Is there a way to substract from the resulting mask before saving it? In case I want to use it for a single image for example, I guess video would be more complicated
For one image only, you can replace the Load Video node and the Video Combine node by a Load Image node and Save Image Node. If oyu want one single frame from the resulting video you can use any video editor to get it from the video!
it is not working.
