An Evaluation of the Suitability of FPGAs for Embedded Vision Systems

Reconfigurable hardware, in the form of Field Programmable Gate Arrays (FPGAs), is becoming increasingly attractive for digital signal processing problems, including image processing and computer vision tasks. The ability to exploit the parallelism often found in these problems, as well as the ability to support different modes of operation on a single hardware substrate, gives these devices a particular advantage over fixed architecture devices such as serial CPUs and DSPs. Further, development times are substantially shorter than dedicated hardware in the form of Application Specific ICs (ASICs), and small changes to a design can be prototyped in a matter of hours. On the other hand, designing with FPGAs still requires expertise beyond that found in many vision labs today. This paper looks at the advantages and disadvantages of FPGA technology, its suitability for image processing and computer vision tasks, and attempts to suggest some directions for the future.

[1]  Scott Hauck,et al.  Precis: a design-time precision analysis tool , 2002, Proceedings. 10th Annual IEEE Symposium on Field-Programmable Custom Computing Machines.

[2]  Ramin Zabih,et al.  Non-parametric Local Transforms for Computing Visual Correspondence , 1994, ECCV.

[3]  Peter J. Burt.  A pyramid-based front-end processor for dynamic vision applications , 2002 .

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

[5]  Pierre Marchal,et al.  Field-programmable gate arrays , 1999, CACM.

[6]  Laurent Moll,et al.  Real time correlation-based stereo: algorithm, implementations and applications , 1993 .

[7]  W. James MacLean,et al.  Video-rate stereo depth measurement on programmable hardware , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[8]  Peter J. Burt,et al.  A VLSI pyramid chip for multiresolution image analysis , 1992, International Journal of Computer Vision.

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

[10]  David J. Fleet.  Disparity from local weighted phase-correlation , 1994, Proceedings of IEEE International Conference on Systems, Man and Cybernetics.

[11]  Wayne Luk,et al.  Unifying bit-width optimisation for fixed-point and floating-point designs , 2004, 12th Annual IEEE Symposium on Field-Programmable Custom Computing Machines.

[12]  John Woodfill,et al.  Real-time stereo vision on the PARTS reconfigurable computer , 1997, Proceedings. The 5th Annual IEEE Symposium on Field-Programmable Custom Computing Machines Cat. No.97TB100186).

[13]  E. Yao,et al.  A reconfigurable and flexible parallel 3D vision system for a mobile robot , 1993, 1993 Computer Architectures for Machine Perception.

[14]  I. Xilinx.  Virtex series configuration architecture user guide , 2000 .

[15]  Kurt Konolige,et al.  Small Vision Systems: Hardware and Implementation , 1998 .

[16]  Wayne Luk,et al.  Parameterised floating-point arithmetic on FPGAs , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[17]  Keith D. Underwood,et al.  FPGAs vs. CPUs: trends in peak floating-point performance , 2004, FPGA '04.

[18]  Steve Mann,et al.  Using multiple graphics cards as a general purpose parallel computer: applications to computer vision , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[19]  Ruigang Yang,et al.  Multi-resolution real-time stereo on commodity graphics hardware , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[20]  David G. Lowe,et al.  Object recognition from local scale-invariant features , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[21]  Lesley Shannon,et al.  Simplifying the integration of processing elements in computing systems using a programmable controller , 2005, 13th Annual IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM'05).

[22]  M. Caffrey,et al.  Correcting single-event upsets through virtex partial configuration , 2000 .

[23]  Takeo Kanade,et al.  A stereo machine for video-rate dense depth mapping and its new applications , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[24]  Lev Kirischian,et al.  Re-Configurable Parallel Stream Processor with Self-Assembling and Self-Restorable Micro-Architecture , 2004, Parallel Computing in Electrical Engineering, 2004. International Conference on.