Design and validation of a low resource-cost video data processing method for embedded implementation of optical flow extraction

The main goal of the proposed project is to build real-time hardware implementations of Optical Flow methods for tasks like egomotion estimation and/or obstacle avoidance. The current paper presents the theoretical foundations and the development and adaptation of appropriate algorithms needed for the motion detection task and the selection of the most suitable hardware/software design environments that aids the embedded implementation process. Using proper hardware/software co-design techniques, we present the development of a low-power, low resource-cost video data processing algorithm. The developed hardware efficient Optical flow extraction method is validated via software implementation and tested with standard video sequence inputs. We than present the architecture of the same method on an embedded, FPGA-based platform.

[1]  Hiroaki Niitsuma,et al.  High Speed Computation of the Optical Flow , 2005, ICIAP.

[2]  Dah-Jye Lee,et al.  A Fast and Accurate Tensor-based Optical Flow Algorithm Implemented in FPGA , 2007, 2007 IEEE Workshop on Applications of Computer Vision (WACV '07).

[3]  Javier Díaz,et al.  A multi-resolution approach for massively-parallel hardware-friendly optical flow estimation , 2012, J. Vis. Commun. Image Represent..

[4]  Bhagath Pyda,et al.  A novel high speed L-K based optical flow computation , 2010, 2010 International Conference on Communication and Computational Intelligence (INCOCCI).

[5]  Giulio Sandini,et al.  Uncalibrated obstacle detection using normal flow , 2005, Machine Vision and Applications.

[6]  Richard Szeliski,et al.  A Database and Evaluation Methodology for Optical Flow , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[7]  David J. Fleet,et al.  Performance of optical flow techniques , 1994, International Journal of Computer Vision.

[8]  Sylvain Paris,et al.  SimpleFlow: A Non‐iterative, Sublinear Optical Flow Algorithm , 2012, Comput. Graph. Forum.

[9]  Walter Stechele,et al.  High performance FPGA based optical flow calculation using the census transformation , 2009, 2009 IEEE Intelligent Vehicles Symposium.

[10]  Javier Díaz,et al.  FPGA-based real-time optical-flow system , 2006, IEEE Transactions on Circuits and Systems for Video Technology.

[11]  Pedro Cobos Arribas,et al.  FPGA Real Time Lane Departure Warning Hardware System , 2007, EUROCAST.

[12]  Robert Strzodka,et al.  Real-time motion estimation and visualization on graphics cards , 2004, IEEE Visualization 2004.

[13]  Takeo Kanade,et al.  An Iterative Image Registration Technique with an Application to Stereo Vision , 1981, IJCAI.

[14]  Javier Díaz,et al.  Pipelined architecture for real-time cost-optimized extraction of visual primitives based on FPGAs , 2013, Digit. Signal Process..

[15]  Piotr Dudek,et al.  Hardware Implementation of Skeletonization Algorithm for Parallel Asynchronous Image Processing , 2009, J. Signal Process. Syst..

[16]  Zhiguo Cao,et al.  A real-time embedded architecture for SIFT , 2013, J. Syst. Archit..

[17]  Berthold K. P. Horn,et al.  Determining Optical Flow , 1981, Other Conferences.

[18]  Unai Bidarte,et al.  Hardware implementation of optical flow constraint equation using FPGAs , 2005, Comput. Vis. Image Underst..

[19]  Liang-Gee Chen,et al.  Reconfigurable Morphological Image Processing Accelerator for Video Object Segmentation , 2011, J. Signal Process. Syst..

[20]  T.M. McGinnity,et al.  An Investigation into Optical Flow Computation on FPGA Hardware , 2008, 2008 International Machine Vision and Image Processing Conference.

[21]  W. James MacLean,et al.  An Evaluation of the Suitability of FPGAs for Embedded Vision Systems , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.