Well I have an (set of) image(s) and I wish to create silhouettes out of those. Based on the alpha value of the image (alpha is either 0 or 1 no intermediate values).
The idea I had was to create three layers: a white background layer, a black foreground layer and a third top layer with the image.
Then for the top layer select the right image each time and would pick a blend/layer mode so that it "burns" the alpha from the top layer on the bottom layer. Then I could just load all images into the layers, set the blend modes and make them visible one by one.
Is this possible, or would I have to go through the lengthy process of creating a selection, and making a path from this selection and then using that path as object to fill in the "black" layer. (which requires many more steps and with around a 100 images to process I'm not looking forward to that).
So say I have this image (and/or with the white replaced by a transparent region):
And I wish to create the following: