Low-complexity merge candidate decision for fast HEVC encoding

Muchen Li, Keiichi Chono, Satoshi Goto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

This paper presents a low-complexity Merge candidate decision scheme for fast High-Efficiency Video Coding (HEVC) encoding. Merge mode is a new Motion Vector (MV) representation tool that omits signaling of delta MV information of an inter-coded prediction unit (PU) by sending the index of MV to be merged to that inter-coded PU. Although the Merge mode brings significant coding efficiency improvements, the complexity of its Merge candidate decision associated with Rate-Distortion (R-D) cost computation is proportional to the number of Merge candidates. Thus the Merge candidate decision is a complex encoding process. We reveal redundant and unnecessary R-D cost computation for particular Merge candidates through the investigations on correlation between Ground-truth MVs (GMVs) obtained by motion estimation (ME) and Merge candidates as well as on the position occurrences of selected Merge candidates. We propose a low-complexity Merge candidate decision scheme that determines the most favorable Merge candidate from full Merge candidates by using the GMV and the position information of each Merge candidate of an inter-coded PU. Simulation results demonstrate that, relative to HM-9.0 with one Merge candidate (Ref-1), the proposed scheme achieves 0.16%-1.58% B-D rate reduction without increasing encoding time.

Original languageEnglish
Title of host publicationElectronic Proceedings of the 2013 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2013
DOIs
Publication statusPublished - 2013
Event2013 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2013 - San Jose, CA
Duration: 2013 Jul 152013 Jul 19

Other

Other2013 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2013
CitySan Jose, CA
Period13/7/1513/7/19

Keywords

  • HEVC
  • Merge mode
  • Mode decision

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Low-complexity merge candidate decision for fast HEVC encoding'. Together they form a unique fingerprint.

Cite this