Going deeper into third-person action anticipation

Gulu Mbongeni, Sareh Rowlands

Abstract:

Autonomous systems have become more dominant in our lives and
therefore the ability of these sys-tems to predict future plays an
instrumental role to guarantee assistance and safety. As a result, analysing human actions and anticipation in videos is gaining a great deal
of interest in academic re-search. This paper explores and reviews
different deep learning techniques used in third-person action
anticipation to provide an updated view of advancements in this field.
The task of action anticipation is divided into feature extraction and
a predictive model for many architectures. This paper outlines a
project plan for action anticipation in the third person using stepbased activity. We will use several data sets to compare some of these
different architectures based on their prediction accuracy and ability
to predict actions in varying time frames.