A Comparative Study of Registration Methods for RGB-D Video of Static Scenes

Abstract

The use of RGB-D sensors for mapping and recognition tasks in robotics or, in general, for virtual reconstruction has increased in recent years. The key aspect of these kinds of sensors is that they provide both depth and color information using the same device. In this paper, we present a comparative analysis of the most important methods used in the literature for the registration of subsequent RGB-D video frames in static scenarios. The analysis begins by explaining the characteristics of the registration problem, dividing it into two representative applications: scene modeling and object reconstruction. Then, a detailed experimentation is carried out to determine the behavior of the different methods depending on the application. For both applications, we used standard datasets and a new one built for object reconstruction.

An evaluation of real-time RGB-D visual odometry algorithms on mobile devices

“…To this purpose we recorded a new dataset with various luminosity levels and assessed the accuracy as well as the memory and CPU usage of each of the selected algorithm. Although similar benchmarks can be found in the literature for desktop environment [13,31], to the best of our knowledge, this is the first attempt at benchmarking state-of-the-art algorithms on a mobile device equipped with a depth sensor.…”

A robotic platform for customized and interactive rehabilitation of persons with disabilities

“…Although not the main focus of this work, it is worth mentioning that we have integrated previous works Morell-Gimenez et al, 2014) we have developed for robot navigation and mapping into the mobile robotic platform. These techniques allow the robot to build a 3D map using threedimensional data obtained by an RGB-D sensor.…”

Joint bias estimation and localization in factor graph

“…x m = r m cos θ m = (r + r b +r) cos(θ + θ b +θ) =x +x (9) y m = r m sin θ m = (r + r b +r) sin(θ + θ b +θ) =ȳ +ỹ (10) wherex =r cosθ andȳ =r sinθ denote the true values and x andỹ denote the systematic errors.…”