Can someone, who may know the under-the-hood workings, advise -
I have been lead to believe that if an object is removed, using the new-ish AI removal tool, that if the size is over 1024 x 1024 (?) pixels in size (I have also read it's about 2000 x 2000), the result is a lower resolution, so that area looks a bit blurry compared to the rest of the image.
So, two questions - Does this dimension refer to square footage or the longest dimension, i.e. if I try to remove an object that remains within 1024 x 1024 (let's say it is 1000 x 1000) it remains high res, but if it is 100 x 1500 (so extends outside the boundary in one dimension but well within in the other and it is less square footage) it is low res? - And, if I wish to remove a branch as an example, that is well larger than the noted constraint, if I "nibble" at it in small bites, each below the constraint, will the eventual result be high res?
Just curious how this thing works.
