Fast Transform Unit Depth Decision Based on Quantized Coefficients for HEVC

Zhaoqing PAN, Jianjun LEI, Yun ZHANG, Wenjie YAN, Sam KWONG

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

2 Citations (Scopus)

Abstract

The quad tree structure based Transform Unit (TU) helps high efficiency video coding to improve the coding efficiency. However, the achieved coding efficiency comes at the cost of the increased computational complexity. In this paper, based on the quantizated coefficients of the TU, we propose an early termination for the quad tree structure based TU encoding process. If the quantized coefficients of the luminance components are all zeros, the TU encoding process will be terminated. Experimental results show that the proposed method achieves about 55.13% on average TU encoding time saving, while the rate distortion performance degradation is negligible.
Original languageEnglish
Title of host publicationThe 2015 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2015
PublisherIEEE
Pages1127-1132
Number of pages6
ISBN (Electronic)9781479986972
DOIs
Publication statusPublished - Oct 2015
Externally publishedYes
EventIEEE International Conference on Systems, Man, and Cybernetics, SMC 2015 - Kowloon Tong, Hong Kong
Duration: 9 Oct 201512 Oct 2015

Conference

ConferenceIEEE International Conference on Systems, Man, and Cybernetics, SMC 2015
Abbreviated titleSMC 2015
Country/TerritoryHong Kong
CityKowloon Tong
Period9/10/1512/10/15

Keywords

  • fast algorithm
  • HEVC
  • Quadtree
  • transform unit
  • video coding

Fingerprint

Dive into the research topics of 'Fast Transform Unit Depth Decision Based on Quantized Coefficients for HEVC'. Together they form a unique fingerprint.

Cite this