The selection offered by ‘Subject Mask’ is usually very good, but in many cases it isn’t perfect. It’s often necessary to correct this subject selection.
If you combine this ‘corrected’ subject mask with, for example, a radial filter, the correction is not taken into consideration. Only the mask automatically found by the AI is taken into account. It would be nice if you could correct this subject selection and have the corrected mask recognised as the subject.
I’d like to explain my idea in a bit more detail.
When identifying a subject mask, C1P automatically determines what it considers to be the subject. Occasionally, this selection is inaccurate or significantly inaccurate from time to time. A correction is then necessary.
However, when combining a subject mask with other masks, the reference is not made to the corrected subject mask, but to what C1P considers to be the subject – in other words, the uncorrected mask. This is often frustrating, as it means a correction has to be made all over again.
It becomes particularly inconvenient and tedious when there are several combinations of masks that reference the subject mask. It would therefore be very helpful if, for all possible combinations, C1P were to reference the subject mask that has already been created and corrected, rather than calculating a new one each time.
Thanks!
Regards
Martin