A Scene-based Augmented Reality Framework for Exhibits

This paper presents a novel augmented reality (AR) framework specifically targeting scene-based exhibits. Unlike traditional AR libraries that rely on specific image targets, the proposed framework utilizes a bag-of-words model for scene recognition, which enables recognition and subsequent launch of AR experiences under challenging environments. Moreover, the proposed framework utilizes relocalization capabilities of the popular ORB-SLAM algorithm to track the user’s movement after recognition of a particular exhibit. We demonstrate the efficacy of the proposed framework through a complete mobile app designed for a local museum, where artifacts are enhanced through AR.