Tried several different cameras with people, vehicles, animals etc prominently in the shot. PL10’s ‘ai-selection’ button at the base of the masking window whirls for a few seconds & gives me just “subject” or “background” every time.
The object selector will work to select an individual or vehicle (usually) but the auto-selector not.
So I uninstalled the app and carefully cleaned everything labelled “PL 10” from the disk. Restarted. Reinstalled PL10. Same problem.
After PL9, I’m more than a little tired of DXO releases being so buggy.
(MacBook Pro M4, 36GB, latest OS: Nikon + Olympus + Leica raw images tested)
I find similar mixed results with AI auto-selection, although generally it works a little better with people, but isn’t intelligent enough to select e.g. a person and the guiter they’re holding.
For vehicles it’s very spotty. I prefer the rectangular “area select” tool for that, and generally it does a good job.
Draw a box around what you want to select or use the “square box” to select what you want if the AI is not clever enough with the predefined selections like subject, persons etc.
Testing PL10 you should also investigate combining masks and try functions like Invert shape and Invert Mask.
In addition, it’s the Depth mask, a very fine tool.
I for one has never been able to select exactly what I want to select as easily as with PL10, but of course you have to invest a little time to learn how to use it best!
I agree with you that “Invert Shape” is not a good label for the function and that “Exclude Shape” is better, but I feel that whatever labeled is used on the masking functions in PL10 are better than ever and very actually very good.
But of course, plenty of options can make it a bit complex and as a user you have to invest some time learn and use it at it’s best.
Hey, there are not that much functions here. Just the basics for blending masks. And it works fine.
About masks, it would be great to be able to “flatten” combination of masks into a mask alone. It would be great to be able to save masks as images on disk (for other uses) and it would be great to be able to load images from disk as masks.
If we’re talking about improving masks, I’d really like the interface to be larger and for the “sub-folders” of masks to remain expanded.
The way I edit can involve copy/pasting masks or duplicating them and it quickly becomes a pain to move things about because the “sub-folders” keep closing and there’s only a tiny window to work with.
