Abstract
Abstract. The development of Mobile Mapping systems over the last decades allowed to quickly collect georeferenced spatial measurements by means of sensors mounted on mobile vehicles. Despite the large number of applications that can potentially take advantage of such systems, because of their cost their use is currently typically limited to certain specialized organizations, companies, and Universities. However, the recent worldwide diffusion of powerful mobile devices typically embedded with GPS, Inertial Navigation System (INS), and imaging sensors is enabling the development of small and compact mobile mapping systems. More specifically, this paper considers the development of a 3D reconstruction system based on photogrammetry methods for smartphones (or other similar mobile devices). The limited computational resources available in such systems and the users' request for real time reconstructions impose very stringent requirements on the computational burden of the 3D reconstruction procedure. This work takes advantage of certain recently developed mathematical tools (incremental singular value decomposition) and of photogrammetry techniques (structure from motion, Tomasi–Kanade factorization) to access very computationally efficient Euclidian 3D reconstruction of the scene. Furthermore, thanks to the presence of instrumentation for localization embedded in the device, the obtained 3D reconstruction can be properly georeferenced.
Highlights
The developments of photogrammetry during the last decades allowed to obtain high resolution 3D models of the reality from camera measurements by means of well known and typically quite computational demanding signal processing procedures
Some methods have been proposed in the literature to reduce such computational effort: most of them are based on the optimization of bundle adjustment methods, e.g. the use of Preconditioned Conjugate Gradients to speed up the bundle adjustment optimization (Agarwal et al, 2010, Byrod and Astrom, 2010)
This paper proposes the integration of the above factorization algorithm with the information provided by the navigation system embedded in the device in order to obtain a fast and effective georeferenced reconstruction of the Euclidian 3D structure of the scene: thanks to the use of computationally efficient methods, the overall reconstruction procedure can be executed in real time on standard mobile devices, e.g. smartphones
Summary
The developments of photogrammetry during the last decades allowed to obtain high resolution 3D models of the reality from (quite low cost) camera measurements by means of well known and typically quite computational demanding signal processing procedures (data association and structure from motion methods). Of Preconditioned Conjugate Gradients to speed up the bundle adjustment optimization (Agarwal et al, 2010, Byrod and Astrom, 2010) From such methods, this work takes advantage of the Incremental Singular Value Decomposition (ISVD (Brand, 2002)) to obtain a fast factorization of the measurement matrix (Tomasi and Kanade’s factorization (Tomasi and Kanade, 1992)). This procedure, which has been recently proposed by Kennedy et al (Kennedy et al, 2013), allows to quickly obtain a projective reconstruction of the scene. Since the proposed procedure allows to obtain georeferenced 3D reconstructions, the outcomes of this method can be used both as reconstructions for an imaging system, and as feedback information for the navigation system in order to improve its localization
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
