Title :
Parallel, real-time visual SLAM
Author :
Clipp, Brian ; Lim, Jongwoo ; Frahm, Jan-Michael ; Pollefeys, Marc
Author_Institution :
Dept. of Comput. Sci., Univ. of North Carolina, Chapel Hill, NC, USA
Abstract :
In this paper we present a novel system for real-time, six degree of freedom visual simultaneous localization and mapping using a stereo camera as the only sensor. The system makes extensive use of parallelism both on the graphics processor and through multiple CPU threads. Working together these threads achieve real-time feature tracking, visual odometry, loop detection and global map correction using bundle adjustment. The resulting corrections are fed back into to the visual odometry system to limit its drift over long sequences. We demonstrate our system on a series videos from challenging indoor environments with moving occluders, visually homogenous regions with few features, scene parts with large changes in lighting and fast camera motion. The total system performs its task of global map building in real time including loop detection and bundle adjustment on typical office building scale scenes.
Keywords :
SLAM (robots); cameras; computer graphic equipment; coprocessors; distance measurement; feature extraction; image motion analysis; image sequences; multi-threading; multiprocessing systems; robot vision; stereo image processing; video signal processing; camera motion; global map correction; graphics processor; image sequence; loop detection; multiple CPU thread; occluder; parallel real-time visual SLAM; real-time feature tracking; six degree of freedom; stereo camera; video signal processing; visual odometry; visual simultaneous localization and mapping;
Conference_Titel :
Intelligent Robots and Systems (IROS), 2010 IEEE/RSJ International Conference on
Conference_Location :
Taipei
Print_ISBN :
978-1-4244-6674-0
DOI :
10.1109/IROS.2010.5653696