Efficient inter-prediction depth coding algorithm based on depth map segmentation for 3D-HEVC

Yi Wen Liao, Mei Juan Chen, Chia Hung Yeh*, Jie Ru Lin, Chih Wei Chen

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

The 3D extension of High Efficiency Video Coding (3D-HEVC) is the latest international 3D video coding standard, which enhances the compression efficiency of 3D videos with multi-view plus depth (MVD) format. However, it is at the cost of increasing coding complexity. Therefore, this paper proposes a fast inter-prediction algorithm based on depth segmentation for the depth coding of 3D-HEVC to reduce the coding complexity of the depth map. We aim to reduce the coding time by efficiently utilizing the properties of the depth map. First, the proposed algorithm divides a depth map into background, middle ground and foreground, based on automatic thresholding technique. Then, we adjust the search range according to the classification of the coding tree unit (CTU). In addition, an early termination decision by utilizing the correlation of spatial, temporal, and inter-view neighboring CTUs and a fast prediction unit mode decision are also proposed to reduce coding time. The experimental results show that the proposed algorithm can reduce the depth coding time and still maintain good video quality.

Original languageEnglish
Pages (from-to)10181-10205
Number of pages25
JournalMultimedia Tools and Applications
Volume78
Issue number8
DOIs
Publication statusPublished - 2019 Apr 1

Keywords

  • 3D-HEVC
  • Automatic thresholding
  • Depth coding
  • Inter prediction

ASJC Scopus subject areas

  • Software
  • Media Technology
  • Hardware and Architecture
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Efficient inter-prediction depth coding algorithm based on depth map segmentation for 3D-HEVC'. Together they form a unique fingerprint.

Cite this