Skip to main content

Advertisement

SpringerLink
Log in

Fast and energy-efficient approximate motion estimation architecture for real-time 4 K UHD processing

  • Original Research Paper
  • Published:
Journal of Real-Time Image Processing Aims and scope Submit manuscript

Abstract

Approximate computing techniques exploit the characteristics of error-tolerant applications either to provide faster implementations of their computational structures or to achieve substantial improvements in terms of energy efficiency. In video encoding, the motion estimation (ME) stage, including the Integer ME (IME) and the Fractional ME (FME) steps, is the most computational intensive task and it is highly resilient to controlled losses of accuracy. In accordance, this article proposes the exploitation of approximate computing techniques to implement energy efficient dedicated hardware structures targeting the motion estimation stage of current video encoders. The designed ME architecture supports IME and FME and is able to real-time process 4 K UHD videos (3840 × 2160 pixels) at 30 frames per second, while dissipating 108.92 mW. When running at its maximum operation frequency, the architecture can process 8 K UHD videos (7680 × 4320 pixels) at 120 frames per second. The solution described in this article presents the highest throughput and the highest energy efficiency among all state-of-the-art compared works, showing that the use of approximate computing is a promising solution when implementing video encoders in dedicated hardware.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

References

  1. Sullivan, G., Ohm, J., Han, W., Wiegand, T.: Overview of the high efficiency video coding (HEVC) standard. IEEE Trans. Circuits Syst. Video Technol. (2012). https://doi.org/10.1109/TCSVT.2012.2221191

    Article  Google Scholar 

  2. He, Z., Yu, L., Zheng, X., Ma, S.: AVS2-video coding standard - an application-oriented and high performance video coding standard. IEEE Int. Conf. Multimedia Expo Workshops (2014). https://doi.org/10.1109/ICMEW.2014.6890707

    Article  Google Scholar 

  3. AOM: Alliance for Open Media: Get started creating products with AV1. https://aomedia.org/av1-features/get-started/ (2019). Accessed 08 May 2019.

  4. MPEG: The Moving Picture Experts Group: Versatile Video Coding. https://mpeg.chiariglione.org/standards/mpeg-i/versatile-video-coding (2019). Accessed 08 May 2019.

  5. Han, J., Orshansky, M.: Approximate computing: an emerging paradigm for energy-efficient design. In: 18th IEEE European Test Symposium (2013). https://doi.org/10.1109/ETS.2013.6569370.

  6. Chippa, V., Venkataramani, S., Chakradhar, S., Roy, K., Raghunathan, A.: Approximate computing: an integrated hardware approach. Asilomar Conf Signals Syst. Comput. (2013). https://doi.org/10.1109/ACSSC.2013.6810241

    Article  Google Scholar 

  7. Raha, A., Jayakumar, H., Raghunathan, V.: A Power efficient video encoder using reconfigurable approximate arithmetic units. In: 27th International Conference on VLSI Design and 2014 13th International Conference on Embedded Systems (2014). https://doi.org/10.1109/VLSID.2014.62.

  8. Porto, R., Agostini, L., Zatt, B., Porto, M., Roma, N., Sousa, L.: Energy-efficient motion estimation with approximate arithmetic. In: IEEE 19th International Workshop on Multimedia Signal Processing (2017). https://doi.org/10.1109/MMSP.2017.8122248.

  9. Alvar, S., Abdollahzadeh, M., Seyedarabi, H.: A novel fast search motion estimation algorithm in video coding. In: IEEE 23rd International Symposium on Industrial Electronics (2014). https://doi.org/10.1109/ISIE.2014.6864737.

  10. Chiang, J., Kuo, W., Su, L.: Fast motion estimation using hexagon-based search pattern in predictive search range. Int. Conf. Comput. Commun. Netw. (2007). https://doi.org/10.1109/ICCCN.2007.4317974

    Article  Google Scholar 

  11. Jou, S., Chang, S., Chang, T.: Fast motion estimation algorithm and design for real time QFHD high efficiency video coding. IEEE Trans. Circuits Syst. Video Technol. (2015). https://doi.org/10.1109/TCSVT.2015.2389472

    Article  Google Scholar 

  12. He, G., Zhou, D., Li, Y., Chen, Z., Zhang, T., Goto, S.: High-throughput power-efficient VLSI architecture of fractional motion estimation for ultra-HD HEVC video encoding. IEEE Trans. Very Large Scale Integr. Syst. (2015). https://doi.org/10.1109/TVLSI.2014.2386897

    Article  Google Scholar 

  13. Medhat, A., Shalaby, A., Sayed, M.: High-throughput hardware implementation for motion estimation in HEVC encoder. In: IEEE 58th International Midwest Symposium on Circuits and Systems (2015). https://doi.org/10.1109/MWSCAS.2015.7282040.

  14. Park, S., Choi, B., Lim, I., Park, H., Kang, S.: An efficient motion estimation hardware architecture using Modified Reference Data Access (MRDAS) skip algorithm for high Efficiency Video Coding (HEVC) encoder. In: IEEE 6th International Conference on Consumer Electronics – Berlin (2016). https://doi.org/10.1109/ICCE-Berlin.2016.7684724.

  15. Singh, K., Ahamed, S.: Low power motion estimation algorithm and architecture of HEVC/H265 for consumer applications. IEEE Trans. Consum. Elect. (2018). https://doi.org/10.1109/TCE.2018.2867823

    Article  Google Scholar 

  16. Jia, L., Tsui, C., Au, O., Jia, K.: A low-power motion estimation architecture for HEVC based on a new sum of absolute difference computation. IEEE Trans. Circuits Syst. Video Technol. (2018). https://doi.org/10.1109/TCSVT.2018.2890204

    Article  Google Scholar 

  17. Mahdiani, H., Ahmadi, A., Fakhraie, S., Lucas, C.: Bio-inspired imprecise computational blocks for efficient VLSI implementation of soft-computing applications. IEEE Trans. Circuits Syst. I Regul. Pap. (2010). https://doi.org/10.1109/TCSI.2009.2027626

    Article  MathSciNet  Google Scholar 

  18. Porto, R., Agostini, L., Zatt, B., Roma, N., Porto, M.: Power-efficient approximate SAD architecture with LOA imprecise adders. In: IEEE 10th Latin American Symposium on Circuits & Systems (2019). https://doi.org/10.1109/LASCAS.2019.8667554.

  19. Perleberg, M., Afonso, V., Conceição, R., Susin, A., Agostini, L., Porto, M., Zatt, B.: Energy and rate-aware design for HEVC motion estimation based on Pareto efficiency. J. Integrat. Circuits Syst. 13, 1 (2018)

    Article  Google Scholar 

  20. Hwang, S., Ha, J., Sunwoo, M.: Efficient integer motion estimation algorithm using sub-sampling. Int. SoC Design Conf. (2009). https://doi.org/10.1109/SOCDC.2009.5423844

    Article  Google Scholar 

  21. Kang, S., Yoo, D., Lee, S., Kim, Y.: Hardware implementation of motion estimation using a sub-sampled block for frame rate up-conversion. Int. SoC Design Conf. (2008). https://doi.org/10.1109/SOCDC.2008.4815694

    Article  Google Scholar 

  22. Frustaci, F., Lanuzza, M., Zicari, P., Perri, S., Corsonello, P.: Designing high-speed adders in power-constrained environments. IEEE Trans. Circ. Syst. II Express Briefs (2009). https://doi.org/10.1109/TCSII.2008.2010187

    Article  Google Scholar 

  23. Dutt, S., Nandi, S., Trivedi, G.: A comparative survey of approximate adders. In: 26th International Conference Radioelektronika (2016). https://doi.org/10.1109/RADIOELEK.2016.7477392.

  24. Kahng, A., Kang, S.: Accuracy-configurable adder for approximate arithmetic designs. Design Autom. Conf. (2012). https://doi.org/10.1145/2228360.2228509

    Article  Google Scholar 

  25. Camus, V., Schlachter, J., Enz, C.: A low-power carry cut-back approximate adder with fixed-point implementation and floating-point precision. In: 53nd ACM/EDAC/IEEE Design Automation Conference (2016). https://doi.org/10.1145/2897937.2897964.

  26. Zhu, N., Goh, W., Zhang, W., Yeo, K., Kong, Z.: Design of low-power high-speed truncation-error-tolerant adder and its application in digital signal processing. IEEE Trans. Very Large Scale Integr. Syst. (2010). https://doi.org/10.1109/TVLSI.2009.2020591

    Article  Google Scholar 

  27. Zhu, N., Goh, W., Wang, G., Yeo, K.: Enhanced low-power high-speed adder for error-tolerant application. Int. SoC Design Conf. (2010). https://doi.org/10.1109/SOCDC.2010.5682905

    Article  Google Scholar 

  28. Shafique, M., Ahmad, W., Hafiz, R., Henkel, J.: A low latency generic accuracy configurable adder. ACM/EDAC/IEEE Design Autom. Conf. (2015). https://doi.org/10.1145/2744769.2744778

    Article  Google Scholar 

  29. Bossen, F.: Common test conditions and software reference configurations. Document JCTVC-L1100, ITU-T SG16 WP3 and ISO/IEC JTC1/SC29/WG11 Joint Collaborative Team on Video Coding (JCT-VC). (2013).

  30. Bjontegaard, G.: Improvements of the BD-PSNR model, VCEG-AI11, ITU-T SG16/Q6 VCEG 35th meeting, Berlin, Germany, 16–18 (2008).

  31. Silvaco: Nangate FreePDK45 open cell library. https://www.silvaco.com/products/nangate/FreePDK45_Open_Cell_Library/index.html (2019). Accessed 08 May 2019.

  32. High Efficiency Video Coding (HEVC): Reference software. https://hevc.hhi.fraunhofer.de (2019). Accessed 08 May 2019.

  33. Ghosh, A., Dehuri, S.: Evolutionary algorithms for multi-criterion optimization: a survey. Int. J. Comput. Inf. Sci. 2, 38–57 (2004)

    Google Scholar 

  34. Song, C., Ju, L., Jia, Z.: Hybrid scratchpad and cache memory management for energy-efficient parallel HEVC encoding. IEEE Int. Conf. Comput. Design (2015). https://doi.org/10.1109/ICCD.2015.7357185

    Article  Google Scholar 

  35. Chen, C., Huang, C., Chen, Y., Chen, L.: Level C+ data reuse scheme for motion estimation with corresponding coding orders. IEEE Trans. Circuits Syst. Video Technol. (2006). https://doi.org/10.1109/TCSVT.2006.871388

    Article  Google Scholar 

  36. Afonso, V., Maich, H., Audibert, L., Zatt, B., Porto, M., Agostini, L., Susin, A.: Hardware implementation for the HEVC fractional motion estimation targeting real-time and low-energy. J. Integrat. Circuits Syst. 11, 106–120 (2016)

    Article  Google Scholar 

Download references

Acknowledgements

This work is partly financed by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior–Brasil (CAPES) Finance Code 001, by FCT projects PTDC/EEI-HAC/30485/2017 and UIDB/50021/2020, and also by CNPq and FAPERGS Brazilian research support agencies.

Author information

Authors and Affiliations

  1. Video Technology Research Group (ViTech), Group of Architectures and Integrated Circuits (GACI), Graduate Program in Computing (PPGC), Federal University of Pelotas (UFPel), Pelotas, Brazil

    Roger Porto, Murilo Perleberg, Vladimir Afonso, Bruno Zatt, Luciano Agostini & Marcelo Porto

  2. Sul-Rio-Grandense Federal Institute of Science and Technology (IFSul), Pelotas, Brazil

    Roger Porto & Vladimir Afonso

  3. Instituto Superior Técnico (IST), Universidade de Lisboa (ULisboa), Lisboa, Portugal

    Nuno Roma

  4. Instituto de Engenharia de Sistemas e Computadores, Investigação e Desenvolvimento em Lisboa (INESC-ID), Lisboa, Portugal

    Nuno Roma

Authors
  1. Roger Porto
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Murilo Perleberg
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Vladimir Afonso
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Bruno Zatt
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Nuno Roma
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Luciano Agostini
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Marcelo Porto
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Roger Porto.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Porto, R., Perleberg, M., Afonso, V. et al. Fast and energy-efficient approximate motion estimation architecture for real-time 4 K UHD processing. J Real-Time Image Proc 18, 723–737 (2021). https://doi.org/10.1007/s11554-020-01014-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11554-020-01014-6

Keywords

Search

Navigation