Earticle

초록

영어

HEVC video coding standard can achieve an average of 50% bit-rate saving, transmitting nearly the same visual quality, compared with the H.264/AVC High Profile. This paper focuses on the key features of the HEVC standard. Then it analyzes the complexity of the HEVC standard by profiling the HEVC reference software HM12.0 which is released by JCT-VC community with Intel Core i7 processor. The experiment covers different settings for a wide variety of video sequences, with the resolutions ranging from 240p to 1600p. The evaluation result shows that the HEVC standard can achieve a doubling in coding efficiency at the cost of increasing the complexity slightly. When the sequences are coded in all-intra or random-access configurations, intra prediction, loop filter, inverse transform and motion compensation are the most complex components for HEVC decoder. The motion compensation part should be given lots of consideration since it can take nearly half of the decoding time. The quantization parameter (QP) also has obvious influences on the complexity of the decoder. Besides, in practice, the HM decoder is not able to decode video content with 1080p or higher resolutions for real-time applications.

목차

Abstract
 1. Introduction
 2. Key Features of the HEVC Standard
  2.1. More Flexible Block Partitioning and Bigger Blocks
  2.2. Improved Mechanisms to Support Parallel Processing
  2.3. More Intra Prediction Modes for Intra Prediction
  2.4. Improved Motion Information Encoding
  2.5. Improved Non-Integer Pixel Interpolation
  2.6. Better Approximated Transform Coefficients
  2.7. Improved Entropy Coding
  2.8. Alternative Sao Filter
 3. Analysis of HEVC Architecture
  3.1. HEVC Encoder
  3.2. HEVC Decoder
 4. Evalution of HEVC Reference Software
  4.1. Test Conditions and Test Sequences
  4.2. Profiling Results for HM Encoder
  4.3. Profiling Results for HM Decoder
 5. Conclusion
 Acknowledgements
 References

키워드

저자

• Jianjun Wang [ School of Information Science and Engineering, Shandong University, Jinan, Shandong, China ]
• Li Zhou [ School of Information Science and Engineering, Shandong University, Jinan, Shandong, China ]
• Tao Sun [ Shandong Provincial Key Laboratory of Network based Intelligent Computing, University of Jinan, Jinan, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Hybrid Information Technology

• 간기

  격월간

• pISSN

  1738-9968

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605