The emergence of visual-based localization and navigation using smartphone sensing

The research on localization and navigation has a long history. Traditional location-based services do not work well for indoor environments. The reason is that, compared with outdoors, indoor environments are more complicated and GPS signals are not available. Furthermore, existing indoor localization systems either need to deploy extra devices or are not accurate/reliable. In this paper, we present an emerging research area of mobile crowdsensing — visual-based localization and navigation (VBLN). It makes use of citizen's smartphone cameras and other internal sensors to attain indoor localization and navigation, which proves effective and of low deployment cost. We describe and discuss the generic architecture, its characters, typical applications/systems, key challenges/techniques, as well as the open issues of VBLN.