You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm trying to merge annotations of non-overlapping slices (so overlaps in slicing are both set to 0.0). I was under the impression that SAHI could merge adjacent bboxes, e.g. an object split in half between two adjacent slices.
I annotated the object both left and right, but when merging the annotations and using SAHI postprocessing (tried all algorithms and various threshold values) it seems the both annotations remain in the final set.
I also understand that when SAHI predicts on the entire image also, then there may be an encompassing bbox and that will provide overlap between left and right-adjacent bboxes, thus allowing for merging due to bbox overlap.
But does SAHI not support merging adjacent bboxes that have 0.0 overlap in two slices?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
I'm trying to merge annotations of non-overlapping slices (so overlaps in slicing are both set to 0.0). I was under the impression that SAHI could merge adjacent bboxes, e.g. an object split in half between two adjacent slices.
I annotated the object both left and right, but when merging the annotations and using SAHI postprocessing (tried all algorithms and various threshold values) it seems the both annotations remain in the final set.
I also understand that when SAHI predicts on the entire image also, then there may be an encompassing bbox and that will provide overlap between left and right-adjacent bboxes, thus allowing for merging due to bbox overlap.
But does SAHI not support merging adjacent bboxes that have 0.0 overlap in two slices?
Beta Was this translation helpful? Give feedback.
All reactions