High Performance Stereo System for Dense 3-D Reconstruction

3-D stereo reconstruction, a technique that estimates per-pixel depth in a scene, is still a challenging problem mainly due to some prohibitive factors that limit its performance and computational ability. The aim of this paper is to present a new hardware-efficient disparity map computation, which is based on disparity space image processing using discrete dynamic systems. The hardware architecture of the proposed system was implemented on a high-end field programmable gate array (FPGA) device, offering real-time 3-D reconstruction speeds using a hardware aware architecture based on parallelism and process pipelining. The proposed architecture fulfills the requirements of real-world applications regarding resource usage, frame rates, and disparity resolution, while its implementation on an Altera Stratix IV family FPGA device can extract disparity maps of up to 1280 × 1024 pixels with up to 128 disparity levels under real-time or near real-time conditions at a clock rate of 168 MHz. Qualitative and quantitative results also demonstrate its performance and improvement over previous hardware-related studies, making our approach a suitable candidate for applications in which timing and processing constraints are critical.

[1]  Radim Sára,et al.  Finding the Largest Unambiguous Component of Stereo Matching , 2002, ECCV.

[2]  Alberto Prieto,et al.  Real-Time System for High-Image Resolution Disparity Estimation , 2007, IEEE Transactions on Image Processing.

[3]  Yunde Jia,et al.  A miniature stereo vision machine (MSVM-III) for dense disparity mapping , 2004, ICPR 2004.

[4]  Darius Burschka,et al.  Advances in Computational Stereo , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Theocharis Theocharides,et al.  Hardware Design Considerations for Edge-Accelerated Stereo Correspondence Algorithms , 2012, VLSI Design.

[6]  S. Sabihuddin,et al.  Dynamic programming approach to high frame-rate stereo correspondence: A pipelined architecture implemented on a field programmable gate array , 2008, 2008 Canadian Conference on Electrical and Computer Engineering.

[7]  Dan Popovici,et al.  Cellular Automata in Image Processing , 2002 .

[8]  Yoram Yakimovsky,et al.  A system for extracting three-dimensional measurements from a stereo pair of TV cameras , 1976 .

[9]  James D. Meiss,et al.  Differential dynamical systems , 2007, Mathematical modeling and computation.

[10]  Theocharis Theocharides,et al.  Towards hardware stereoscopic 3D reconstruction a real-time FPGA computation of the disparity map , 2010, 2010 Design, Automation & Test in Europe Conference & Exhibition (DATE 2010).

[11]  Heiko Hirschmüller,et al.  Improvements in real-time correlation-based stereo vision , 2001, CVPR 2001.

[12]  Tsutomu Maruyama,et al.  A Real-Time Stereo Vision System with FPGA , 2003, FPL.

[13]  JunSeong Kim,et al.  Real-Time Stereo Vision on a Reconfigurable System , 2005, SAMOS.

[14]  Harlan Hile,et al.  Stereo Video Processing for Depth Map , 2004 .

[15]  Jiangbo Lu,et al.  High-Speed Dense Stereo Via Directional Center-Biased Windows on Graphics Hardware , 2007, 2007 3DTV Conference.

[16]  Peter Pirsch,et al.  Real-time stereo vision system using semi-global matching disparity estimation: Architecture and FPGA-implementation , 2010, 2010 International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation.

[17]  John Iselin Woodfill,et al.  Tyzx DeepSea High Speed Stereo Vision System , 2004, 2004 Conference on Computer Vision and Pattern Recognition Workshop.

[18]  Giulio Sandini,et al.  Learning mobile robot navigation: a behavior-based approach , 1994, Proceedings of IEEE International Conference on Systems, Man and Cybernetics.

[19]  Sek M. Chai,et al.  Multi-Resolution Real-Time Dense Stereo Vision Processing in FPGA , 2012, 2012 IEEE 20th International Symposium on Field-Programmable Custom Computing Machines.

[20]  Richard Szeliski,et al.  A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms , 2001, International Journal of Computer Vision.

[21]  Miguel Arias-Estrada,et al.  Multiple Stereo Matching Using an Extended Architecture , 2001, FPL.

[22]  N. Felber,et al.  Efficient ASIC implementation of a real-time depth mapping stereo vision system , 2003, 2003 46th Midwest Symposium on Circuits and Systems.

[23]  J.M. Perez,et al.  High memory throughput FPGA architecture for high-definition Belief-Propagation stereo matching , 2009, 2009 3rd International Conference on Signals, Circuits and Systems (SCS).

[24]  Yasuhiro Kobayashi,et al.  FPGA implementation of a stereo matching processor based on window-parallel-and-pixel-parallel architecture , 2005, 48th Midwest Symposium on Circuits and Systems, 2005..

[25]  Andreas Steininger,et al.  SAD-Based Stereo Matching Using FPGAs , 2009 .

[26]  Alexander M. Bronstein,et al.  Embedded System for 3 D Shape Reconstruction , 2008 .

[27]  W. James MacLean,et al.  A Real-Time Large Disparity Range Stereo-System Using FPGAs , 2006, ACCV.

[28]  James D. Meiss.  Differential Dynamical Systems (Monographs on Mathematical Modeling and Computation) , 2007 .

[29]  Antonios Gasteratos,et al.  An Algorithm for Adaptive Mean Filtering and Its Hardware Implementation , 2006, J. VLSI Signal Process..

[30]  A. C. Sonmez,et al.  FPGA design and implementation of a real-time stereo vision system , 2012, 2012 International Symposium on Innovations in Intelligent Systems and Applications.