Towards Next Best View Planning for Time-Variant Scenes

Abstract

In modelling and reconstruction of objects or scenes, dynamic settings present different challenges than the more well-studied static case. In this work, we explore these challenges by investigating how next best view planning can be adapted for observation of dynamic scenes. We conduct a thorough review of different representations of information based on precedent from the static case, and find that view planning cannot be directly transferred from static environments to dynamic scenes without accounting for information deterioration and observational bias.