tracking programatically using a bounding box (or central point) as input?

[aendrs]

Let's say that I have a set of videos in which I want to track an object and get masks in return. As an input I have a bounding box containing the single object to be tracked, from which I can also derive the center of the box in case a single point is preferred. Is there a straightforward way to do it programmatically, without the manual user interface?

Thanks for the feedback

[RaulKite]

Any advance on this?

Thanks

[RunsenXu]

I have the same question here. Actually, we can use the bounding box or center point to prompt sam to get the mask at the first frame. But usually, the mask won't be perfect sufferfing from over-seg or under-seg. What would be a good solution for this? Thanks!