Processing Geotagged Image Sets for Collaborative Compositing and View Construction

In this paper we present a method for local processing of photos and associated sensor information on mobile devices. Our goal is to lay the foundations of a collaborative multi-user framework where ad-hoc device groups can share their data around a geographical location to produce more complex composited views of the area, without the need of a centralized server-client - cloud-based - architecture. We focus on processing as much data locally on the devices as possible, and reducing the amount of data that needs to be shared. The main results are the proposal of a lightweight processing and feature extraction framework, based on the analysis of vision graphs, and presenting preliminary composite view generation based on these results.