*Local Editing > Smoothing*

I’ve been studying the @carafife video on Smoothing. In this video the Area Mask is used to subtract the eyes from the Parametric mask. About 19:25 into the video. https://www.youtube.com/watch?v=SZXTLC-s1SE

However, this is what happens when I activate on the Area Mask in order to subtract from the Parametric mask

Mask created

Area mask in subtract mode to remove this shudder from mask. Everything is now subtracted.

If I select the shudder that I want to keep in add mode, then the other areas are removed from the mask. This behavior seems to be the opposite of what would be expected where I would normally select the area to be subtracted.

I’m probably missing something, but I would like to know how to achieve what the author was able to do in the way that he did it, if possible

You need to add a first area rectangle (in normal mode) covering the whole image, and then subtract from it with other shapes.
The subtraction mode is relative to other shapes in the area mask. Then the whole area is always intersected with other masks.
The only exception to this is the brush, which when set to add mode can be used to modify any other mask.

Hope this helps

1 Like

Thanks, Alberto, that does help. A lot!

Since you are not the first one to be confused by this, I was thinking of adding a special case to the area mask, that adds automatically a rectangle covering the whole image if the first shape is in subtract mode.
What do people think about this? I usually don’t like special cases, because their behaviour can easily get confusing, but I’d still like to hear some other opinions…

Based on my original question, how would this then relate to creating a mask in Local Edits > Smoothing and then subtracting from the mask? Not sure I understand.

masks are created in the same way in all the tools supporting local editing. In your case, if you want to subtract something from a parametric mask, you do:

  1. create your parametric mask
  2. create an area mask with a rectangle in normal mode, covering the whole picture
  3. add another rectangle in subtract mode, to exclude the areas you don’t want to modify

my question above was whether it makes sense to have a “special case” that would avoid the user to explicitly perform step 2.
Hope it’s clearer now…

Yes, avoiding the need for step 2 would definitely make sense as long as there are other scenarios where this would create issues. Otherwise, why not.

Would this also work with Color Similarity?

yes, same thing

1 Like

If I understand correctly, from a practical point of view, it makes sense to me to have the rectangle covering the screen if the first shape is in subtract mode. Without it you are subtracting a shape from nothing so nobody would want to do that. I can also see why it sort of goes against the grain…

I think it could be more practical as drawing a mask covering the whole image in order to subtract a region is not intuitive.

1 Like

The only iffy scenario I can think of is when the user starts off by creating a full image mask (spot) and then decides to subtract from the mask. Would this mean that two full image masks would have be created, one by the user and one when activating subtract?

I thought more about this, and my idea does more harm than good. It will easily get confusing. So, not going to change the behaviour

Step #2 is not a big deal. Once done, your already there to subtract whatever you don’t want included the mask.

One idea is if there are subtract and/or interest masks with no add mask the software could alert the user that the mask won’t do anything. But I’m not sure if there is a mechanism place to notify the user about anything right now. If not that’s a whole new addition and may not be worth the trouble.


I like this. Here it is:

1 Like

Looks good. I think that will be helpful.

Just passing by so take this with appropriate (dis)regard. Usability wise it would be better to disable (grey out) substract/interest masks until there are masks to substract from. Should be a second stage of GUI.

1 Like