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
When there are tiny floating supervoxels that aren't connected to the main body, the cleave server assigns them a label of 0. Neu3 then warns the user that some supervoxels remained unassigned (#227).
However, a common scenario is that many tiny pieces remain unassigned, and hunting them down is tricky. Furthermore, they aren't really important enough to hunt down and seed separately. We should just auto-seed them each with a unique label when the user commits the cleave.
That is, if N supervoxels remain unaassigned when the user commits a cleave, should just be blasted apart into N independent bodies. Since the number will often be greater than 20, perhaps it can't really be visualized. It can only be done at the very end.
[BTW, there's a chance that this is already the behavior of neu3 in such cases. I just don't remember. If that's true, then sorry for the unnecessary feature request.]
When there are tiny floating supervoxels that aren't connected to the main body, the cleave server assigns them a label of 0. Neu3 then warns the user that some supervoxels remained unassigned (#227).
However, a common scenario is that many tiny pieces remain unassigned, and hunting them down is tricky. Furthermore, they aren't really important enough to hunt down and seed separately. We should just auto-seed them each with a unique label when the user commits the cleave.
That is, if N supervoxels remain unaassigned when the user commits a cleave, should just be blasted apart into N independent bodies. Since the number will often be greater than 20, perhaps it can't really be visualized. It can only be done at the very end.
[BTW, there's a chance that this is already the behavior of neu3 in such cases. I just don't remember. If that's true, then sorry for the unnecessary feature request.]
Related: #125, #227, #228.
cc @knechtc @hubbardp @tingzhao
The text was updated successfully, but these errors were encountered: