Robust Real-Time Visual Odometry for Dense RGB-D Mapping - Robotics Institute Carnegie Mellon University

Robust Real-Time Visual Odometry for Dense RGB-D Mapping

Thomas Whelan, Hordur Johannsson, Michael Kaess, John J. Leonard, and John McDonald
Conference Paper, Proceedings of (ICRA) International Conference on Robotics and Automation, pp. 5724 - 5731, May, 2013

Abstract

This paper describes extensions to the Kintinuous [1] algorithm for spatially extended KinectFusion, incorporating the following additions: (i) the integration of multiple 6DOF camera odometry estimation methods for robust tracking; (ii) a novel GPU-based implementation of an existing dense RGB-D visual odometry algorithm; (iii) advanced fused real-time surface coloring. These extensions are validated with ex- tensive experimental results, both quantitative and qualitative, demonstrating the ability to build dense fully colored models of spatially extended environments for robotics and virtual reality applications while remaining robust against scenes with challenging sets of geometric and visual features.

BibTeX

@conference{Whelan-2013-7720,
author = {Thomas Whelan and Hordur Johannsson and Michael Kaess and John J. Leonard and John McDonald},
title = {Robust Real-Time Visual Odometry for Dense RGB-D Mapping},
booktitle = {Proceedings of (ICRA) International Conference on Robotics and Automation},
year = {2013},
month = {May},
pages = {5724 - 5731},
}