A high speed iterative closest point tracker on an FPGA platform

This paper presents and examines a hardware implementation of a high speed Iterative Closest Point (ICP) based object tracking system, which uses stereo vision disparities as input. Custom field programmable gate array (FPGA) hardware has been designed to handle the inherent bottlenecks that result from the large input and processing bandwidths of the range data. The custom hardware has been implemented and tested on various objects, using both software simulation and hardware tests. Results indicate that the tracker is able to successfully track freeform objects along arbitrary paths at rates of over 200 frames-per-second. Tracking errors are low, in spite of substantial sensor and stereo extraction noise. The tracker is able to track linear paths within 1.57 mm and 2.80 degrees and gracefully degrades under occlusion. This high speed hardware implementation with 16 parallel nearest neighbor units has a five times speed improvement when compared to a software k-d tree implementation.

[1]  Hong Li,et al.  Segmentation and Recognition of Continuous Gestures , 2007, 2007 IEEE International Conference on Image Processing.

[2]  P. Jasiobedzki,et al.  Model based pose estimation for autonomous operations in space , 1999, Proceedings 1999 International Conference on Information Intelligence and Systems (Cat. No.PR00446).

[3]  F. A. Seiler,et al.  Numerical Recipes in C: The Art of Scientific Computing , 1989 .

[4]  Martin Jägersand,et al.  3D SSD tracking with estimated 3D planes , 2009, Image Vis. Comput..

[5]  Frank Dellaert,et al.  Robust car tracking using Kalman filtering and Bayesian templates , 1998, Other Conferences.

[6]  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..

[7]  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..

[8]  J. Little,et al.  Satellite Pose Acquisition and Tracking with Variable Dimensional Local Shape Descriptors , 2005 .

[9]  Babak Taati,et al.  Variable Dimensional Local Shape Descriptors for Object Recognition in Range Data , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[10]  Michael A. Greenspan,et al.  Efficient tracking with the Bounded Hough Transform , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[11]  J. L. Roux.  An Introduction to the Kalman Filter , 2003 .

[12]  Gérard G. Medioni,et al.  Object modelling by registration of multiple range images , 1992, Image Vis. Comput..

[13]  Heinrich Niemann,et al.  Performance Analysis of Nearest Neighbor Algorithms for ICP Registration of 3-D Point Sets , 2003, VMV.

[14]  Mark Fiala,et al.  ARTag, a fiducial marker system using digital techniques , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[15]  P. Abbeel,et al.  Kalman filtering , 2020, IEEE Control Systems Magazine.

[16]  Luc Van Gool,et al.  Object Tracking with an Adaptive Color-Based Particle Filter , 2002, DAGM-Symposium.

[17]  Katsuhiko Sakaue,et al.  Registration and integration of multiple range images for 3-D model construction , 1996, Proceedings of 13th International Conference on Pattern Recognition.

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

[19]  Jack Ritter.  A fast approximation to 3D Euclidian distance , 1990 .

[20]  Marc Levoy,et al.  Zippered polygon meshes from range images , 1994, SIGGRAPH.

[21]  Gerhard Roth,et al.  Pose Determination and Tracking for Autonomous Satellite Capture , 2001 .

[22]  Marc Rioux,et al.  Three-dimensional registration using range and intensity information , 1994, Other Conferences.

[23]  Martin D. Levine,et al.  Registering Multiview Range Data to Create 3D Computer Objects , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[24]  Berthold K. P. Horn,et al.  Closed-form solution of absolute orientation using unit quaternions , 1987 .

[25]  Michael A. Greenspan,et al.  Pose Determination By PotentialWell Space Embedding , 2007, Sixth International Conference on 3-D Digital Imaging and Modeling (3DIM 2007).

[26]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[27]  Marc Levoy,et al.  Efficient variants of the ICP algorithm , 2001, Proceedings Third International Conference on 3-D Digital Imaging and Modeling.

[28]  Takeo Kanade,et al.  Real-time 3-D pose estimation using a high-speed range sensor , 1993, Proceedings of the 1994 IEEE International Conference on Robotics and Automation.

[29]  Christophe Bobda,et al.  A Dynamic Reconfigurable Hardware/Software Architecture for Object Tracking in Video Streams , 2006, EURASIP J. Embed. Syst..

[30]  Michael A. Greenspan,et al.  The parallel iterative closest point algorithm , 2001, Proceedings Third International Conference on 3-D Digital Imaging and Modeling.

[31]  Weiwei Ma,et al.  An FPGA-Based Singular Value Decomposition Processor , 2006, 2006 Canadian Conference on Electrical and Computer Engineering.

[32]  W. James MacLean,et al.  Maximum-Likelihood Stereo Correspondence using Field Programmable Gate Arrays , 2007 .

[33]  Christophe Bobda,et al.  Singular value decomposition on distributed reconfigurable systems , 2001, Proceedings 12th International Workshop on Rapid System Prototyping. RSP 2001.