Title :
Double window optimisation for constant time visual SLAM
Author :
Strasdat, Hauke ; Davison, Andrew J. ; Montiel, J.M.M. ; Konolig, Kurt
Author_Institution :
Dept. of Comput., Imperial Coll. London, London, UK
Abstract :
We present a novel and general optimisation framework for visual SLAM, which scales for both local, highly accurate reconstruction and large-scale motion with long loop closures. We take a two-level approach that combines accurate pose-point constraints in the primary region of interest with a stabilising periphery of pose-pose soft constraints. Our algorithm automatically builds a suitable connected graph of keyposes and constraints, dynamically selects inner and outer window membership and optimises both simultaneously. We demonstrate in extensive simulation experiments that our method approaches the accuracy of offline bundle adjustment while maintaining constant-time operation, even in the hard case of very loopy monocular camera motion. Furthermore, we present a set of real experiments for various types of visual sensor and motion, including large scale SLAM with both monocular and stereo cameras, loopy local browsing with either monocular or RGB-D cameras, and dense RGB-D object model building.
Keywords :
SLAM (robots); image colour analysis; image sensors; pose estimation; robot vision; stereo image processing; RGB-D cameras; constant time visual SLAM; dense RGB-D object model building; double window optimisation; monocular cameras; pose-point constraints; pose-pose soft constraints; reconstruction; stereo cameras; visual sensor; Barium; Cameras; Measurement; Optimization; Simultaneous localization and mapping; Three dimensional displays; Visualization;
Conference_Titel :
Computer Vision (ICCV), 2011 IEEE International Conference on
Conference_Location :
Barcelona
Print_ISBN :
978-1-4577-1101-5
DOI :
10.1109/ICCV.2011.6126517