Anticipating Next Active Objects for Egocentric Videos

arXiv:2302.06358v5 Announce Type: replace
Abstract: This paper addresses the problem of anticipating the next-active-object location in the future, for a given egocentric video clip where the contact might happen, before any action takes place. The problem is considerably hard, as we aim at estimating the position of such objects in a scenario where the observed clip and the action segment are separated by the so-called ``time to contact'' (TTC) segment. Many methods have been proposed to anticipate the action of …

abstract aim arxiv clip cs.cv future location next object objects paper type video video clip videos