Tile Partitioning and Allocation for HEVC Parallel Decoding on Asymmetric Multicores 


Vol. 43,  No. 5, pp. 791-800, May  2018
10.7840/kics.2018.43.5.791


PDF
  Abstract

This paper proposes new HEVC Tile allocation method for asymmetric multicores. The method provides real-time ultra high definition video decoding on mobile devices by using HEVC tiles parallel decoding. To reduce decoding time, it applies novel tile allocation method considering the computational ability of asymmetric multicores as well as the computational complexity of each Tile. In the proposed tile allocation, the complexity of each tile is determined by the number of partitions of the prediction unit (PU) in a tile. The method sums the complexity of each tile and calculates how to allocate to each cores. The complexities have to in proportion to performance ratio of each core. It makes Idle time of each core rarely, and reduces total decoding time. The experimental results show the decoding time speed up from 9 to 16% under JCT-VC common test condition (CTC).

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

H. Roh, B. Lee, E. Ryu, "Tile Partitioning and Allocation for HEVC Parallel Decoding on Asymmetric Multicores," The Journal of Korean Institute of Communications and Information Sciences, vol. 43, no. 5, pp. 791-800, 2018. DOI: 10.7840/kics.2018.43.5.791.

[ACM Style]

Hyun-Joon Roh, Bok-Gi Lee, and Eun-Seok Ryu. 2018. Tile Partitioning and Allocation for HEVC Parallel Decoding on Asymmetric Multicores. The Journal of Korean Institute of Communications and Information Sciences, 43, 5, (2018), 791-800. DOI: 10.7840/kics.2018.43.5.791.

[KICS Style]

Hyun-Joon Roh, Bok-Gi Lee, Eun-Seok Ryu, "Tile Partitioning and Allocation for HEVC Parallel Decoding on Asymmetric Multicores," The Journal of Korean Institute of Communications and Information Sciences, vol. 43, no. 5, pp. 791-800, 5. 2018. (https://doi.org/10.7840/kics.2018.43.5.791)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880