For example, if I wanted to transform part of an image to black and white and leave a section in color, I have been using the Automask tool. A very nice tool, but no matter what the tolerance settings with a moderately complex image, it usually requires returning to the mask/selection and either adding or subtracting to get the precise mask/selection.
So, here's the sweep of the tool across a flower. Even though the green foliage frame is larger than the screencap shown, the circle/adjustment cannot be made any larger on this mouse movement (if that means anything to the mask/selection).
And here's the screen image after the mouse button has been released. The yellow arrow shows the motion of the mouse and the orange circles indicate errors in the mask/selection that have to be fixed with additional "+/-" applications of the Automask tool. (The tolerance on this was set at 0, but other tolerance settings were no better.)
I assume I am doing something wrong here, because this kind of masking/selection should be very quick.
A number of competitors are now using what they call Smart Selection (masking) for this kind of operation. The tool make an intelligent interpretation of what the user wants based on mouse movement. Below is example of a selection/mask made in less than two seconds. The yellow line and arrow show the motion of the mouse to achieve this. No additional tool application was required to achieve this perfect mask/selection.
So, if I can't do this kind of operation in PL as efficiently, I'm faced with the decision about whether to use an alternative editor for masking/selection in my image work
Hopefully, someone can give me the key to quicker automasking/selections in PL.
Thanks.
Bob
