Designing a Posture Analysis System with Hardware Implementation
- First Online:
- 144 Downloads
Posture analysis is an active research area in computer vision for applications such as home care and security monitoring. This paper describes the design of a system for posture analysis with hardware acceleration, addressing the following four aspects: (a) a design workflow for posture analysis based on radial shape and projection histogram representations; (b) the implementation of different architectures based on a high-level hardware design approach with support for automating transformations to improve parallelism and resource optimisation; (c) accuracy evaluation of the proposed posture analysis system, and (d) performance evaluation for the derived designs. One of the designs, which targets a Xilinx XC2V6000 FPGA at 90.2 MHz, is able to perform posture analysis at a rate of 1,164 frames per second with a frame size of 320 by 240 pixels. It represents 3.5 times speedup over optimised software running on a 2.4 GHz AMD Athlon 64 3700+ computer. The frame rate is well above that of real-time video, which enables the sharing of the FPGA among multiple video sources.
Keywordsposture analysis gait analysis hardware compilation FPGA ubiquitous sensor networks
- 1.Advanced Micro Devices (AMD) Inc., http://www.amd.com.
- 2.A. Azarbayejani, C. Wren, and A. Pentland, “Real-time 3D Tracking of the Human Body,” in Proc. of IMAGE’COM 96, 1996.Google Scholar
- 3.T. Boult, “Frame-rate Multibody Tracking for Surveillance,” in Proc. of DARPA Image Understanding Workshop, 1998.Google Scholar
- 4.Celoxica Ltd, http://www.celoxica.com/.
- 5.C. C. Cheung, W. Luk, and P. Y. K. Cheung, “Reconfigurable Elliptic Curve Cryptosystem on a Chip,” in Proc. Int. Conf. on Design Automation and Test in Europe (DATE), vol. 1, 2005, pp. 24–29.Google Scholar
- 6.J. G. F. Coutinho, J. Jiang, and W. Luk, “Interleaving Behavioural and Cycle-accurate Descriptions for Reconfigurable Hardware Compilation,” in IEEE Symposium on Field-Programmable Custom Computing Machines, 2005.Google Scholar
- 9.E. Grimson and C. Stauffer, “Adaptive Background Mixture Models for Real Time Tracking,” in Proc. of the Computer Vision and Pattern Recognition Conference, 1999.Google Scholar
- 11.Intel Corporation, http://www.intel.com.
- 12.M. P. T. Juvonen, J. G. F. Coutinho, J. L. Wang, B. L. Lo, W. Luk, O. Mencer, and G. Z. Yang, “Custom Hardware Architectures for Posture Analysis,” in IEEE International Conference on Field Prog. Tech., 2005.Google Scholar
- 13.A. Lipton, H. Fujiyoshi, and H. Patil, “Moving Target Detection and Classification from Real-time Video,” in Proc. of the IEEE Workshop Application of Computer Vision, 1998.Google Scholar
- 14.B. Lo, J. L. Wang, and G. Z. Yang, “From Imaging Networks to Behavior Profiling: Ubiquitous Sensing for Managed Homecare of the Elderly,” in Adjunct Proc. of the 3rd International Conference on Pervasive Computing, May 2005.Google Scholar
- 15.W. Luk, T. K. Lee, J. R. Rice, P. Y. K. Cheung, and N. Shirazi, “Reconfigurable Computing for Augmented Reality,” in Proc. of the IEEE Symposium on Field-Programmable Custom Computing Machines, 1999, pp. 136–145.Google Scholar
- 17.T. Olson and F. Brill, “Moving Object Detection and Event Recognition Algorithms for Smart Cameras,” in Proc. of DARPA Image Understanding Workshop, 1997, pp. 159–175.Google Scholar
- 18.J. M. Rehg, M. Loughlin, and K. Waters, “Vision for a Smart Kiosk,” in IEEE Conference Computer Vision and Pattern Recognition, 1997.Google Scholar
- 20.TriMedia TM1300, http://www.tm1300.com/.
- 22.J. Villasenor, B. Schoner, K. Chia, and C. Zapata, “Configurable Computing Solutions for Automatic Target Recognition,” in Proc. IEEE Symposium on FPGAs for Custom Computing Machines, 1996, pp. 70–79.Google Scholar
- 24.C. Wren, A. Azarbayejani, T. Darrell, and A. Pentland, “Pfinder: Real-time Tracking of the Human Body,” in Pfinder: Real-time Tracking of the Human Body, vol. 19, no. 7, 1997, pp. 780–785.Google Scholar