This paper describes a novel sensor system to estimate the motion of a stereo camera. Local invariant image features are matched between pairs of frames and linked into image trajectories at video rate, providing the so-called visual odometry, i.e., motion estimates from visual input alone. Our proposal conducts two matching sessions: the first one between sets of features associated to the images of the stereo pairs and the second one between sets of features associated to consecutive frames. With respect to previously proposed approaches, the main novelty of this proposal is that both matching algorithms are conducted by means of a fast matching algorithm which combines absolute and relative feature constraints. Finding the largest-valued set of mutually consistent matches is equivalent to finding the maximum-weighted clique on a graph. The stereo matching allows to represent the scene view as a graph which emerge from the features of the accepted clique. On the other hand, the frame-to-frame matching defines a graph whose vertices are features in 3D space. The efficiency of the approach is increased by minimizing the geometric and algebraic errors to estimate the final displacement of the stereo camera between consecutive acquired frames. The proposed approach has been tested for mobile robotics navigation purposes in real environments and using different features. Experimental results demonstrate the performance of the proposal, which could be applied in both industrial and service robot fields.
References
[1]
Zaman, M. High resolution relative localisation using two cameras. Rob. Autonomous Syst 2007, 55, 685–692.
[2]
Nister, D; Naroditsky, O; Bergen, J. Visual Odometry. Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’04), Washington, DC, USA, 27 June–2 July 2004. Volume 1; pp. 652–659.
[3]
Konolige, K; Agrawal, M. Frame-Frame Matching for Realtime Consistent Visual Mapping. Proceedings of the IEEE International Conference on Robotics and Automation, ICRA 2007, Roma, Italy, 10–14 April 2007; pp. 2803–2810.
[4]
Klein, G; Murray, D. Improving the agility of keyframe-based SLAM. Lect. Note Computer. Sci 2008, 5303, 802–815.
[5]
Cheng, Y; Maimone, MW; Matthies, L. Visual odometry on the Mars exploration rovers—a tool to ensure accurate driving and science imaging. IEEE Rob. Autom. Mag 2006, 13, 54–62.
[6]
Johnson, AE; Goldberg, SB; Cheng, Y; Matthies, L. Robust and Efficient Stereo Feature Tracking for Visual Odometry. Proceedings of the IEEE International Conference on Robotics and Automation, ICRA 2008, Pasadena, CA, USA, 19–23 May 2008; pp. 39–46.
[7]
Pretto, A; Menegatti, E; Bennewitz, M; Burgard, W; Pagello, E. A visual Odometry Framework Robust to Motion Blur. Proceedings of the IEEE International Conference on Robotics and Automation, ICRA 2009, Kobe, Japan, 12–17 May 2009; pp. 2250–2257.
[8]
Howard, A. Real-Time Stereo Visual Odometry for Autonomous Ground Vehicles. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, Nice, France, 22–26 September 2008; pp. 3946–3952.
[9]
Hirschmuller, H; Innocent, PR; Garibaldi, JM. Fast, unconstrained camera motion estimation from stereo without tracking and robust statistics. Proceedings of 7th International Conference on Control, Automation, Robotics and Vision, ICARCV 2002, Singapore, 2–5 December 2002; pp. 1099–1104.
[10]
Hartley, R; Zisserman, A. Multiple View Geometry in Computer Vision, 2nd ed ed.; Cambridge University Press: Cambridge, UK, 2004.
[11]
Lowe, DG. Object Recognition from Local Scale-Invariant Features. Proceedings of the International Conference on Computer Vision, Corfu, Greece, September 1999; pp. 1150–1157.
[12]
Bay, H; Tuytelaars, T; Gool, LV. SURF: Speeded Up Robust Features. Proceedings of the 9th European Conference on Computer Vision, Graz, Austria, 7–13 May 2006.
[13]
Harris, C; Stephens, MJ. A Combined Corner and Edge Detector. Proceedings of the Fourth Alvey Vision Conference, Manchester, UK, 31 August–2 September 1988; pp. 147–152.
[14]
Barrow, HG; Burstall, RM. Subgraphs isomorphism, matching relational structures and maximal cliques. Inf. Process. Lett 1976, 4, 83–84.
[15]
Se, S; Lowe, D; Little, J. Mobile robot localization and mapping with uncertainty using scale invariant visual landmark. Int. J. Rob. Res 2002, 21, 735–758.
[16]
Kumlander, D. On Importance of a Special Sorting in the Maximum-Weight Clique Algorithm Based on Colour Classes. Proceedings of the 2nd International Conference Modelling, Computation and Optimization in Information Systems, Metz, France, 8–10 September 2008; pp. 165–174.
[17]
？stergard, P. A new algorithm for the maximum-weight clique problem. Electron. Notes Discrete Math 1999, 3, 153–156.
[18]
Alismail, H; Browning, B; Dias, MB. Evaluating Pose Estimation Method for Stereo Visual Odometry on Robots. Proceedings of the 6th International Conference on Intelligent Autonomous Systems, ICAS 2010, Cancun, Mexico, 7–13 March 2010.
[19]
Umeyama, S. Least-squares estimation of transformation parameters between two point patterns. IEEE Trans. Pattern Anal. Mach. Intell 1991, 13, 376–380.
[20]
Fischler, MA; Bolles, RC. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography. Commun. ACM 1981, 24, 381–395.
[21]
Beis, J; Lowe, DG. Shape Indexing Using Approximate Nearest-Neighbour Search in High-Dimensional Spaces. Proceedings of the Conference on Computer Vision and Pattern Recognition, San Juan, Puerto Rico, 17–19 June 1997; pp. 1000–1006.
[22]
Núnez, P; Vázquez-Martín, R; Bandera, A; Sandoval, F. Combined Constraint Matching Algorithm for Stereo Visual Odometry Based on Local Interest Points. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2009, St. Louis, MO, USA, 10–15 October 2009; pp. 3095–3100.
[23]
Wu, C. SiftGPU: A GPU Implementation of Scale Invariant Feature Transform (SIFT). 2007. Available online: http://www.cs.unc.edu/ccwu/siftgpu/ (accessed on 14 July 2011).
[24]
Blanco, JL; González, J; Fernández-Madrigal, JA. An Experimental Comparison of Image Feature Detectors and Descriptors applied to Grid Map Matching. Technical Report; University of Málaga: Malaga, Spain, 2010.
[25]
Technologies of Vision [Intelligent Computer Vision], EU Projects. Available online: http://tev.fbk.eu (accessed on 14 July 2011).
[26]
Nú？ez, P; Vázquez-Martín, R; Bandera, A; Sandoval, F. Fast laser scan matching approach based on adaptive curvature estimation for mobile robots. Robotica 2009, 27, 469–479.
[27]
Motzkin, TS; Straus, EG. Maxima for graphs and a new proof of a theorem of Turán. Can. J. Math 1965, 17, 533–540.
