Suppose a ball is rolling down an incline as follows:

enter image description here

In the middle of the path, a curtain blocks the camera view of the ball momentarily. Soon the ball reappears after passing behind the curtain.

How will a vision transformer (causal attention over time) applied to a video sequence know whether to draw the black ball or not after it dissapears behind the curtain?



You must log in to answer this question.