A feature matching method for simultaneous localization and mapping

Simultaneous Localization and Mapping is an important technology which help a mobile robot to determine its location and build the environment map. Recently, the RGBD sensor is widely used in the robot, research on RGBD-SLAM becomes a hot topic. In order to calculate the movement parameters of robot, feature matching is adopted to register the two adjacent RGBD images in the video stream. This paper proposed an improved feature matching method for RGBD-SLAM. The experiment results show that, compared with the traditional SIFT feature matching methods for RGBD-SLAM, the performance of the proposed method is improved significantly.

[1]  Wolfram Burgard,et al.  An evaluation of the RGB-D SLAM system , 2012, 2012 IEEE International Conference on Robotics and Automation.

[2]  Luc Van Gool,et al.  Speeded-Up Robust Features (SURF) , 2008, Comput. Vis. Image Underst..

[3]  James J. Little,et al.  Vision-based mobile robot localization and mapping using scale-invariant features , 2001, Proceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (Cat. No.01CH37164).

[4]  Derek Hoiem,et al.  Indoor Segmentation and Support Inference from RGBD Images , 2012, ECCV.

[5]  J. M. M. Montiel,et al.  ORB-SLAM: A Versatile and Accurate Monocular SLAM System , 2015, IEEE Transactions on Robotics.

[6]  Pierre Vandergheynst,et al.  FREAK: Fast Retina Keypoint , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[7]  Christopher G. Harris,et al.  A Combined Corner and Edge Detector , 1988, Alvey Vision Conference.