A Dynamic Reconfigurable Hardware/Software Architecture for Object Tracking in Video Streams

This paper presents the design and implementation of a feature tracker on an embedded reconfigurable hardware system. Contrary to other works, the focus here is on the efficient hardware/software partitioning of the feature tracker algorithm, a viable data flow management, as well as an efficient use of memory and processor features. The implementation is done on a Xilinx Spartan 3 evaluation board and the results provided show the superiority of our implementation compared to the other works.

[1]  Jürgen Teich,et al.  The Erlangen Slot Machine: a highly flexible FPGA-based reconfigurable platform , 2005, 13th Annual IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM'05).

[2]  Heinrich Niemann,et al.  Efficient Feature Tracking for Long Video Sequences , 2004, DAGM-Symposium.

[3]  Pietro Perona,et al.  Real Time Motion Detection System and Scene Segmentation , 1998 .

[4]  James R. Bergen,et al.  Visual odometry , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[5]  Majid Sarrafzadeh,et al.  Reconfiguration in network of embedded systems: Challenges and adaptive tracking case study , 2005, J. Embed. Comput..

[6]  Stefano Soatto,et al.  Fast Visual Feature Selection and Tracking in a Hybrid Reconfigurable Architecture , 2004 .

[7]  Howard Jay Siegel,et al.  SIMD and mixed-mode implementations of a visual tracking algorithm , 1998, Proceedings of the First Merged International Parallel Processing Symposium and Symposium on Parallel and Distributed Processing.

[8]  Carlo Tomasi,et al.  Good features to track , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Howard Jay Siegel,et al.  Implementations of a feature-based visual tracking algorithm on two MIMD machines , 1997, Proceedings of the 1997 International Conference on Parallel Processing (Cat. No.97TB100162).

[10]  J. Chen.  Real Time Feature Extraction and Tracking in a Computational Steering Environment , 2003 .

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