FDQM

Fast Quality Metric for Depth Maps Without View Synthesis

Won Dong Jang, Tae Young Chung, Jae Young Sim, Chang-Su Kim

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

We propose a fast quality metric for depth maps, called fast depth quality metric (FDQM), which efficiently evaluates the impacts of depth map errors on the qualities of synthesized intermediate views in multiview video plus depth applications. In other words, the proposed FDQM assesses view synthesis distortions in the depth map domain, without performing the actual view synthesis. First, we estimate the distortions at pixel positions, which are specified by reference disparities and distorted disparities, respectively. Then, we integrate those pixel-wise distortions into an FDQM score by employing a spatial pooling scheme, which considers occlusion effects and the characteristics of human visual attention. As a benchmark of depth map quality assessment, we perform a subjective evaluation test for intermediate views, which are synthesized from compressed depth maps at various bitrates. We compare the subjective results with objective metric scores. Experimental results demonstrate that the proposed FDQM yields highly correlated scores to the subjective ones. Moreover, FDQM requires at least 10 times less computations than conventional quality metrics, since it does not perform the actual view synthesis.

Original languageEnglish
Article number6963368
Pages (from-to)1099-1112
Number of pages14
JournalIEEE Transactions on Circuits and Systems for Video Technology
Volume25
Issue number7
DOIs
Publication statusPublished - 2015 Jul 1

Fingerprint

Pixels

Keywords

  • 3-D video
  • depth map quality assessment
  • image quality assessment
  • multiview video plus depth (MVD)
  • spatial pooling
  • virtual view synthesis

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Media Technology

Cite this

FDQM : Fast Quality Metric for Depth Maps Without View Synthesis. / Jang, Won Dong; Chung, Tae Young; Sim, Jae Young; Kim, Chang-Su.

In: IEEE Transactions on Circuits and Systems for Video Technology, Vol. 25, No. 7, 6963368, 01.07.2015, p. 1099-1112.

Research output: Contribution to journalArticle

Jang, Won Dong ; Chung, Tae Young ; Sim, Jae Young ; Kim, Chang-Su. / FDQM : Fast Quality Metric for Depth Maps Without View Synthesis. In: IEEE Transactions on Circuits and Systems for Video Technology. 2015 ; Vol. 25, No. 7. pp. 1099-1112.
@article{16ffcc23f0c2434b9a07bf3c830934ef,
title = "FDQM: Fast Quality Metric for Depth Maps Without View Synthesis",
abstract = "We propose a fast quality metric for depth maps, called fast depth quality metric (FDQM), which efficiently evaluates the impacts of depth map errors on the qualities of synthesized intermediate views in multiview video plus depth applications. In other words, the proposed FDQM assesses view synthesis distortions in the depth map domain, without performing the actual view synthesis. First, we estimate the distortions at pixel positions, which are specified by reference disparities and distorted disparities, respectively. Then, we integrate those pixel-wise distortions into an FDQM score by employing a spatial pooling scheme, which considers occlusion effects and the characteristics of human visual attention. As a benchmark of depth map quality assessment, we perform a subjective evaluation test for intermediate views, which are synthesized from compressed depth maps at various bitrates. We compare the subjective results with objective metric scores. Experimental results demonstrate that the proposed FDQM yields highly correlated scores to the subjective ones. Moreover, FDQM requires at least 10 times less computations than conventional quality metrics, since it does not perform the actual view synthesis.",
keywords = "3-D video, depth map quality assessment, image quality assessment, multiview video plus depth (MVD), spatial pooling, virtual view synthesis",
author = "Jang, {Won Dong} and Chung, {Tae Young} and Sim, {Jae Young} and Chang-Su Kim",
year = "2015",
month = "7",
day = "1",
doi = "10.1109/TCSVT.2014.2372343",
language = "English",
volume = "25",
pages = "1099--1112",
journal = "IEEE Transactions on Circuits and Systems for Video Technology",
issn = "1051-8215",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "7",

}

TY - JOUR

T1 - FDQM

T2 - Fast Quality Metric for Depth Maps Without View Synthesis

AU - Jang, Won Dong

AU - Chung, Tae Young

AU - Sim, Jae Young

AU - Kim, Chang-Su

PY - 2015/7/1

Y1 - 2015/7/1

N2 - We propose a fast quality metric for depth maps, called fast depth quality metric (FDQM), which efficiently evaluates the impacts of depth map errors on the qualities of synthesized intermediate views in multiview video plus depth applications. In other words, the proposed FDQM assesses view synthesis distortions in the depth map domain, without performing the actual view synthesis. First, we estimate the distortions at pixel positions, which are specified by reference disparities and distorted disparities, respectively. Then, we integrate those pixel-wise distortions into an FDQM score by employing a spatial pooling scheme, which considers occlusion effects and the characteristics of human visual attention. As a benchmark of depth map quality assessment, we perform a subjective evaluation test for intermediate views, which are synthesized from compressed depth maps at various bitrates. We compare the subjective results with objective metric scores. Experimental results demonstrate that the proposed FDQM yields highly correlated scores to the subjective ones. Moreover, FDQM requires at least 10 times less computations than conventional quality metrics, since it does not perform the actual view synthesis.

AB - We propose a fast quality metric for depth maps, called fast depth quality metric (FDQM), which efficiently evaluates the impacts of depth map errors on the qualities of synthesized intermediate views in multiview video plus depth applications. In other words, the proposed FDQM assesses view synthesis distortions in the depth map domain, without performing the actual view synthesis. First, we estimate the distortions at pixel positions, which are specified by reference disparities and distorted disparities, respectively. Then, we integrate those pixel-wise distortions into an FDQM score by employing a spatial pooling scheme, which considers occlusion effects and the characteristics of human visual attention. As a benchmark of depth map quality assessment, we perform a subjective evaluation test for intermediate views, which are synthesized from compressed depth maps at various bitrates. We compare the subjective results with objective metric scores. Experimental results demonstrate that the proposed FDQM yields highly correlated scores to the subjective ones. Moreover, FDQM requires at least 10 times less computations than conventional quality metrics, since it does not perform the actual view synthesis.

KW - 3-D video

KW - depth map quality assessment

KW - image quality assessment

KW - multiview video plus depth (MVD)

KW - spatial pooling

KW - virtual view synthesis

UR - http://www.scopus.com/inward/record.url?scp=84936774635&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84936774635&partnerID=8YFLogxK

U2 - 10.1109/TCSVT.2014.2372343

DO - 10.1109/TCSVT.2014.2372343

M3 - Article

VL - 25

SP - 1099

EP - 1112

JO - IEEE Transactions on Circuits and Systems for Video Technology

JF - IEEE Transactions on Circuits and Systems for Video Technology

SN - 1051-8215

IS - 7

M1 - 6963368

ER -