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
Is there a way to allocate more grid points as new areas enter in the scene with new frames and other points/track disappear without relying on multiple inference passes on the full sequence?
The text was updated successfully, but these errors were encountered:
I think that the wrong assumption is that the camera is tracking only dynamics targets. But it could be also that camera is moving and you are tracking non-targets. In this case new appearing areas are impacting more the tracking then in the case of dynamic targets and at some point in time you could have too few (or nothing at all) original points to track.
it could nice to have an off the shell option to this case without making complex multiple passes inferences.
Is there a way to allocate more grid points as new areas enter in the scene with new frames and other points/track disappear without relying on multiple inference passes on the full sequence?
The text was updated successfully, but these errors were encountered: