Taming the Memory Demand Complexity of Adaptive Vision Algorithms

  • Majid Sabbagh
  • Hamed Tabkhi
  • Gunar Schirner
Conference paper
Part of the IFIP Advances in Information and Communication Technology book series (IFIPAICT, volume 523)


With the demand for utilizing Adaptive Vision Algorithms (AVAs) in embedded devices, serious challenges have been introduced to vision architects. AVAs may produce huge model data traffic while continuously training the internal model of the stream. This traffic dwarfs the streaming data traffic (e.g. image frames), and consequently dominates bandwidth and power requirements posing great challenges to a low-power embedded implementation. In result, current approaches either ignore targeting AVAs, or are limited to low resolutions due to not handling the traffics separately. This paper proposes a systematic approach to tackle the architectural complexity of AVAs. The main focus of this paper is to manage the huge model data updating traffic of AVAs by proposing a shift from compressing streaming data to compressing the model data. The compression of model data results in significant reduction of memory accesses leading to a pronounced gain in power and performance. This paper also explores the effect of different class of compression algorithms (lossy and lossless) on both bandwidth reduction and result quality of AVAs. For the purpose of exploration this paper focuses on example of Mixture-of-Gaussians (MoG) background subtraction. The results demonstrate that a customized lossless algorithm can maintain the quality while reducing the bandwidth demand facilitating efficient embedded realization of AVAs. In our experiments we achieved the total bandwidth saving of about 69% by applying the Most Significant Bits Selection and BZIP as the first and second level model data compression schemes respectively, with only about 15% quality loss according to the Multi-Scale Structural Similarity (MS-SSIM) metric. The bandwidth saving would be increased to 75% by using a custom compressor.


  1. 1.
    Tabkhi, H., Sabbagh, M., Schirner, G.: Power-efficient real-time solution for adaptive vision algorithms. IET Comput. Digit. Tech. 16–26 (2015)CrossRefGoogle Scholar
  2. 2.
    Xu, J., et al.: A case study in networks-on-chip design for embedded video. In: Design, Automation and Test in Europe (DATE), vol. 2, pp. 770–775 (2004)Google Scholar
  3. 3.
    Lv, T., et al.: A methodology for architectural design of multimedia multiprocessor SoCs. IEEE Des. Test Comput. 22(1), 18–26 (2005)CrossRefGoogle Scholar
  4. 4.
    Chen, G., et al.: Energy savings through compression in embedded Java environments. In: International Symposium on Hardware/Software Codesign, CODES 2002 (2002)Google Scholar
  5. 5.
    Stauffer, C., Grimson, W.E.L.: Adaptive background mixture models for real-time tracking. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 246–252 (1999)Google Scholar
  6. 6.
    Ratnayake, K., Amer, A.: Embedded architecture for noise-adaptive video object detection using parameter-compressed background modeling. J. Real-Time Image Process. (2014)CrossRefGoogle Scholar
  7. 7.
    Xilinx: Programming vision applications on Zynq using OpenCV and high-level synthesis. Xilinx technical report (2013)Google Scholar
  8. 8.
    Tang, Z., Shen, D.: Canny edge detection codec using VLib on Davinci series DSP. In: 2012 International Conference on Computer Science Service System (CSSS) (2012)Google Scholar
  9. 9.
    Swaminathan, K., Lakshminarayanan, G., Ko, S.-B.: High speed generic network interface for network on chip using ping pong buffers. In: International Symposium on Electronic System Design (ISED), pp. 72–76 (2012)Google Scholar
  10. 10.
    Gerstlauer, A., Dömer, R., Peng, J., Gajski, D.D.: System Design: A Practical Guide with SpecC. Kluwer Academic Publisher, Dordrecht (2001)Google Scholar
  11. 11.
    Wang, Z., et al.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)CrossRefGoogle Scholar
  12. 12.
  13. 13.
    Lo, S.-C., Krasner, B., Mun, S.: Noise impact on error-free image compression. IEEE Trans. Med. Imaging (1990)Google Scholar
  14. 14.
    Cosman, P., Gray, R., Olshen, R.: Evaluating quality of compressed medical images: SNR, subjective rating, and diagnostic accuracy. Proc. IEEE (1994)Google Scholar
  15. 15.
    Melnychuck, P.W., Barry, M.J., Mathieu, M.S.: Effect of noise and MTF on the compressibility of high-resolution color images (1990)Google Scholar
  16. 16.
    Pin - a dynamic binary instrumentation tool. Accessed 30 Aug 2015

Copyright information

© IFIP International Federation for Information Processing 2017

Authors and Affiliations

  1. 1.Department of Electrical and Computer EngineeringNortheastern UniversityBostonUSA

Personalised recommendations