Efficient compositional approaches for real-time robust direct visual odometry from RGB-D data
Authors:
- 1 November 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21530858,p. 1100-1106
Abstract
In this paper we give an evaluation of different methods for computing frame-to-frame motion estimates for a moving RGB-D sensor, by means of aligning two images using photometric error minimization. These kind of algorithms have recently shown to be very accurate and robust and therefore provide an attractive solution for robot ego-motion estimation and navigation. We demonstrate three different alignment strategies, namely the Forward-Compositional, the Inverse-Compositional and the Efficient Second-Order Minimization approach, in a general robust estimation framework. We further show how estimating global affine illumination changes, in general improves the performance of the algorithms. We compare our results with recently published work, considered as state-of-the art in this field, and show that our solutions are in general more precise and can perform in real-time on standard hardware.Keywords
This publication has 23 references indexed in Scilit:
- An evaluation of the RGB-D SLAM system2012
- RGB-D mapping: Using Kinect-style depth cameras for dense 3D modeling of indoor environmentsThe International Journal of Robotics Research, 2012
- An asymmetric real-time dense visual localisation and mapping system2011
- DTAM: Dense tracking and mapping in real-time2011
- KinectFusion2011
- KinectFusion: Real-time dense surface mapping and tracking2011
- Dense visual mapping of large scale environments for real-time localisation2011
- View-based MapsThe International Journal of Robotics Research, 2010
- Real-time Quadrifocal Visual OdometryThe International Journal of Robotics Research, 2010
- A Constant-Time Efficient Stereo SLAM System2009