Skip to main content
SpringerLink
Log in

Adaptive data-driven parallelization of multi-view video coding on multi-core processor

  • Published:
Science in China Series F: Information Sciences Aims and scope Submit manuscript

Abstract

Multi-view video coding (MVC) comprises rich 3D information and is widely used in new visual media, such as 3DTV and free viewpoint TV (FTV). However, even with mainstream computer manufacturers migrating to multi-core processors, the huge computational requirement of MVC currently prohibits its wide use in consumer markets. In this paper, we demonstrate the design and implementation of the first parallel MVC system on Cell Broadband Engine™ processor which is a state-of-the-art multi-core processor. We propose a task-dispatching algorithm which is adaptive data-driven on the frame level for MVC, and implement a parallel multi-view video decoder with modified H.264/AVC codec on real machine. This approach provides scalable speedup (up to 16 times on sixteen cores) through proper local store management, utilization of code locality and SIMD improvement. Decoding speed, speedup and utilization rate of cores are expressed in experimental results.

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.

Similar content being viewed by others

References

  1. Smolic A, Kauff P. Interactive 3-D video representation and coding technologies. Proceedings of the IEEE, 2005, 93: 98–110

    Article  Google Scholar 

  2. Dodgson N A, Wiseman N E, Lang S R, et al. Autostereoscopic 3D Displays. IEEE Comput, 2005, 38(8): 31–36

    Google Scholar 

  3. Tanimoto M. FTV (free viewpoint television) creating ray-based image engineering. In: ICIP, 2005

  4. Wilburn B, Joshi N, Vaish V, et al. High performance imaging using large camera arrays. ACM Trans Graph, 2005, 24(3): 765–776

    Article  Google Scholar 

  5. ISO/IEC JTC1/SC29/WG11, Updated Call for Proposals on Multi-view Video Coding, MPEG document N7567, Nice, France, 2005

    Google Scholar 

  6. Vetro A, Wang Y K, Pandit P, et al. Text of ISO/IEC 14496-10:2008/FDAM 1 Multiview Video Coding, MPEG document N9978, Hannover, Germany, 2008

    Google Scholar 

  7. Vetro A, Pandit P, Kimata H, et al. Study Text of ISO/IEC 14496-10:200X/PDAM 1 Multiview Video Coding, MPEG document N9445, Shenzhen, China. 2007

    Google Scholar 

  8. Martinian E, Behrens A, Xin J, et al. Extensions of H.264/AVC for multi-view video compression. In: ICIP, 2006

  9. Shen K, Delp E J. A parallel implementation of an MPEG1 encoder: Faster than real-time! In: SPIE CDVCAT, 1995

  10. Yung H C, Leung K K. Spatial and temporal data parallelization of the H.261 video coding algorithm. IEEE Trans Circuits Sys Video Tech, 2001, 11(1): 91–104

    Article  Google Scholar 

  11. Erik B T, Egbert G J. Mapping of MPEG-4 decoding on a flexible architecture platform. In: SPIE Media Processor, 2002, 4674: 1–13

    Google Scholar 

  12. Erik B T, Egbert G J, Gelderblom R H. Mapping of H.264 decoding on a multiprocessor architecture. In: IVCP, 2003

  13. Wang H, Mao X, Yu L. A novel HDTV decoder and decentralized control scheme. IEEE Trans Consumer Electr, 2001, 47(4): 723–728

    Article  Google Scholar 

  14. Yang Y, Jiang G, Yu M, et al. Parallel process of hyper-space-based multi-view video compression. In: ICIP, 2006

  15. Pang Y, Sun L F, Guo S L, et al. Spatial and temporal data parallelization of multi-view video encoding algorithm. In: MMSP, 2007

  16. Nanda A K, Moulic J R, Hanson R E, et al. Cell/B.E. blades: Building blocks for scalable, real-time, interactive, and digital media servers. IBM J Res & Dev, 2007, 51(5): 573–582

    Google Scholar 

  17. Vetro A, Pandit P, Kimata H, et al. JVT-V209, Joint Draft 2.0 on Multi-view Video Coding. In: MPEG 22nd Meeting, 2007

  18. Kahle J A, Day M N, Hofstee H P, et al. Introduction to the cell multiprocessor. IBM Sys J, 2005, 49(4): 589–604

    Article  Google Scholar 

  19. Flachs B, Asano S, Dhong S H, et al. A streaming processing unit for a CELL processor. In: ISSCC, 2005

  20. MGschwind M, Hofstee H P, Flachs B, et al. A novel SIMD architecture for the cell heterogeneous chip multiprocessor. In: Hot Chips 17 Conference, 2005

  21. Hofstee H P. Introduction to the cell broadband engine. 2005. www-01.ibm.com

  22. Kawada R. KDDI multiview video sequences for MPEG 3DAV use, MPEG document M10533, Munich, Germany, 2004

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science and Technology, Tsinghua University, Beijing, 100084, China

    Yi Pang, WeiDong Hu, LiFeng Sun & ShiQiang Yang

Authors
  1. Yi Pang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. WeiDong Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. LiFeng Sun
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. ShiQiang Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Yi Pang.

Additional information

Supported partially by the National Natural Science Foundation of China (Grant No. 60503063) and the National High-Tech Research & Development Program of China (Grant No. 2006AA01Z321), and the National Basic Research Program of China (Grant No. 2006CB303103)

Rights and permissions

Reprints and permissions

About this article

Cite this article

Pang, Y., Hu, W., Sun, L. et al. Adaptive data-driven parallelization of multi-view video coding on multi-core processor. Sci. China Ser. F-Inf. Sci. 52, 195–205 (2009). https://doi.org/10.1007/s11432-009-0042-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11432-009-0042-8

Keywords

Search

Navigation