A Brute Force Approach to Depth Camera Odometry

By providing direct access to 3D information of the environment, depth cameras are particularly useful for perception applications such as Simultaneous Localization And Mapping or object recognition. With the introduction of the Kinect in 2010, Microsoft released a low cost depth camera that is now intensively used by researchers, especially in the field of indoor robotics. This paper introduces a new 3D registration algorithm that can deal with considerable sensor motion. The proposed approach is designed to take advantage of the powerful computational scalability of Graphics Processing Units (GPU).

[1]  H. Kage,et al.  High speed 3-D registration using GPU , 2008, 2008 SICE Annual Conference.

[2]  Jack Bresenham,et al.  Algorithm for computer control of a digital plotter , 1965, IBM Syst. J..

[3]  Joachim Hertzberg,et al.  Cached k-d tree search for ICP algorithms , 2007, Sixth International Conference on 3-D Digital Imaging and Modeling (3DIM 2007).

[4]  Michael A. Greenspan,et al.  Approximate k-d tree search for efficient ICP , 2003, Fourth International Conference on 3-D Digital Imaging and Modeling, 2003. 3DIM 2003. Proceedings..

[5]  Guido M. Cortelazzo,et al.  3D registration by textured spin-images , 2005, Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05).

[6]  Nico Blodow,et al.  Aligning point cloud views using persistent feature histograms , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[7]  Jun Chu,et al.  Multi-view point clouds registration and stitching based on SIFT feature , 2011, 2011 3rd International Conference on Computer Research and Development.

[8]  Leopoldo Armesto,et al.  A generalization of the metric-based Iterative Closest Point technique for 3D scan matching , 2010, 2010 IEEE International Conference on Robotics and Automation.

[9]  Kazunori Ohno,et al.  Real-Time Robot Trajectory Estimation and 3D Map Construction using 3D Camera , 2006, 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[10]  Hongdong Li,et al.  The 3D-3D Registration Problem Revisited , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[11]  Andreas Nüchter,et al.  Robust 3D-mapping with time-of-flight cameras , 2009, 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[12]  R. Dillmann,et al.  Range Image Registration Using an Octree based Matching Strategy , 2007, 2007 International Conference on Mechatronics and Automation.

[13]  A. Nuchter,et al.  6D SLAM with approximate data association , 2005, ICAR '05. Proceedings., 12th International Conference on Advanced Robotics, 2005..

[14]  Yong Woon Park,et al.  Real-time 3D registration of stereo-vision based range images using GPU , 2009, 2009 Workshop on Applications of Computer Vision (WACV).

[15]  Andreas Nüchter,et al.  GPU-Accelerated Nearest Neighbor Search for 3D Registration , 2009, ICVS.

[16]  Heinz Hügli,et al.  A multi-resolution ICP with heuristic closest point search for fast and robust 3D registration of range images , 2003, Fourth International Conference on 3-D Digital Imaging and Modeling, 2003. 3DIM 2003. Proceedings..

[17]  Fredrik Kahl,et al.  A brute-force algorithm for reconstructing a scene from two projections , 2011, CVPR 2011.

[18]  Hongbin Zha,et al.  Registration of Multiple Laser Scans Based on 3D Contour Features , 2006, Tenth International Conference on Information Visualisation (IV'06).

[19]  Kazufumi Kaneda,et al.  Softassign and EM-ICP on GPU , 2010, 2010 First International Conference on Networking and Computing.

[20]  Paul J. Besl,et al.  A Method for Registration of 3-D Shapes , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Jakob Wasza,et al.  Real-time RGB-D mapping and 3-D modeling on the GPU using the random ball cover data structure , 2011, 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops).