Cheers. Just curious: What made you decide to use G’MIC in the place of ImageMagick? If you need any help with IM, just ask @snibgo. He is on the IM forums too. Both are useful but nowadays I use G’MIC most of the time.
gmic is much faster to run some nice 3D animations (the calculations are done natively) unlike Imagemagick or you had to write a lot of code. A good example is the 3d cube, impossible to create directly with Imagemagick, you must write yourself all the calculations of position of the faces (or use script ex: Fred's ImageMagick Scripts: 3DBOX). It’s a lot longer to run …
I want to avoid using Imagemagick as much as possible when I’m using gmic to improve the speed of the final render.
For the benefit of people searching an answer to this, here’s another answer to this.
s.. c,-{s#-2-1} image[-1] [-3],.5~,.5~,0,0,1,[-2],255 k.
First, split the last channel. Then apply the original[-2] image on top of -1 image with the information found with the image that was separated. I’m pretty sure this can work with CMYKA.
EDIT:
How would I position the center of image into the 0,0 of the target image?
EDIT #2:
The solution to above would be
s.. c,-{s#-2-1} image[-1] [-3],-{w#-3/2},-{h#-3/2},0,0,1,[-2],255 k.
Now, I think I know how to make another paint net plugin for g’mic. I have something in mind.