3-D Video Generation from Monocular Video Based on Hierarchical Video Segmentation

Gwo-Giun Lee, Chun Fu Chen, He Yuan Lin, Ming Jiun Wang

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper presents an algorithm that is capable of generating 3-D video from monocular video through a hierarchical approach that characterizes the low-level features and high-level semantics of video content to generate depth map. Color and texture features are used to locally indicate video’s characteristics followed by segmentation of the video into several regions. Subsequently, semantics, including shape and motion semantics, of segmented regions are delineated from a higher perspective to refine segmentation result via measuring the interrelations among segmented regions. Subsequently, according to refined segmentation map and semantics of segmented regions, the proposed method generates a depth map with high stability using both spatial and temporal information. The stable depth map minimized visual quality degradation, such as flicker and blurring when viewing the 3-D video. The experimental results show the capability of the proposed algorithm in generating a high quality and stable depth map. In addition, subjective viewing evaluation shows that the proposed algorithm surpasses those of the commercial products for converting 2-D video to 3-D video, including TriDef 3D and CyberLink Power DVD.

Original languageEnglish
Pages (from-to)345-358
Number of pages14
JournalJournal of Signal Processing Systems
Volume81
Issue number3
DOIs
Publication statusPublished - 2015 Dec 1

Fingerprint

Video Segmentation
3D Video
Depth Map
Stable Map
Semantics
Segmentation
Subjective Evaluation
Texture Feature
Videodisks
Degradation
Textures
Motion
Color
Experimental Results

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Theoretical Computer Science
  • Signal Processing
  • Information Systems
  • Modelling and Simulation
  • Hardware and Architecture

Cite this

Lee, Gwo-Giun ; Chen, Chun Fu ; Lin, He Yuan ; Wang, Ming Jiun. / 3-D Video Generation from Monocular Video Based on Hierarchical Video Segmentation. In: Journal of Signal Processing Systems. 2015 ; Vol. 81, No. 3. pp. 345-358.
@article{a52d9d35fbee4f079c4df93cfc5172bd,
title = "3-D Video Generation from Monocular Video Based on Hierarchical Video Segmentation",
abstract = "This paper presents an algorithm that is capable of generating 3-D video from monocular video through a hierarchical approach that characterizes the low-level features and high-level semantics of video content to generate depth map. Color and texture features are used to locally indicate video’s characteristics followed by segmentation of the video into several regions. Subsequently, semantics, including shape and motion semantics, of segmented regions are delineated from a higher perspective to refine segmentation result via measuring the interrelations among segmented regions. Subsequently, according to refined segmentation map and semantics of segmented regions, the proposed method generates a depth map with high stability using both spatial and temporal information. The stable depth map minimized visual quality degradation, such as flicker and blurring when viewing the 3-D video. The experimental results show the capability of the proposed algorithm in generating a high quality and stable depth map. In addition, subjective viewing evaluation shows that the proposed algorithm surpasses those of the commercial products for converting 2-D video to 3-D video, including TriDef 3D and CyberLink Power DVD.",
author = "Gwo-Giun Lee and Chen, {Chun Fu} and Lin, {He Yuan} and Wang, {Ming Jiun}",
year = "2015",
month = "12",
day = "1",
doi = "10.1007/s11265-014-0955-3",
language = "English",
volume = "81",
pages = "345--358",
journal = "Journal of Signal Processing Systems",
issn = "1939-8018",
publisher = "Springer New York",
number = "3",

}

3-D Video Generation from Monocular Video Based on Hierarchical Video Segmentation. / Lee, Gwo-Giun; Chen, Chun Fu; Lin, He Yuan; Wang, Ming Jiun.

In: Journal of Signal Processing Systems, Vol. 81, No. 3, 01.12.2015, p. 345-358.

Research output: Contribution to journalArticle

TY - JOUR

T1 - 3-D Video Generation from Monocular Video Based on Hierarchical Video Segmentation

AU - Lee, Gwo-Giun

AU - Chen, Chun Fu

AU - Lin, He Yuan

AU - Wang, Ming Jiun

PY - 2015/12/1

Y1 - 2015/12/1

N2 - This paper presents an algorithm that is capable of generating 3-D video from monocular video through a hierarchical approach that characterizes the low-level features and high-level semantics of video content to generate depth map. Color and texture features are used to locally indicate video’s characteristics followed by segmentation of the video into several regions. Subsequently, semantics, including shape and motion semantics, of segmented regions are delineated from a higher perspective to refine segmentation result via measuring the interrelations among segmented regions. Subsequently, according to refined segmentation map and semantics of segmented regions, the proposed method generates a depth map with high stability using both spatial and temporal information. The stable depth map minimized visual quality degradation, such as flicker and blurring when viewing the 3-D video. The experimental results show the capability of the proposed algorithm in generating a high quality and stable depth map. In addition, subjective viewing evaluation shows that the proposed algorithm surpasses those of the commercial products for converting 2-D video to 3-D video, including TriDef 3D and CyberLink Power DVD.

AB - This paper presents an algorithm that is capable of generating 3-D video from monocular video through a hierarchical approach that characterizes the low-level features and high-level semantics of video content to generate depth map. Color and texture features are used to locally indicate video’s characteristics followed by segmentation of the video into several regions. Subsequently, semantics, including shape and motion semantics, of segmented regions are delineated from a higher perspective to refine segmentation result via measuring the interrelations among segmented regions. Subsequently, according to refined segmentation map and semantics of segmented regions, the proposed method generates a depth map with high stability using both spatial and temporal information. The stable depth map minimized visual quality degradation, such as flicker and blurring when viewing the 3-D video. The experimental results show the capability of the proposed algorithm in generating a high quality and stable depth map. In addition, subjective viewing evaluation shows that the proposed algorithm surpasses those of the commercial products for converting 2-D video to 3-D video, including TriDef 3D and CyberLink Power DVD.

UR - http://www.scopus.com/inward/record.url?scp=84943360139&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943360139&partnerID=8YFLogxK

U2 - 10.1007/s11265-014-0955-3

DO - 10.1007/s11265-014-0955-3

M3 - Article

AN - SCOPUS:84943360139

VL - 81

SP - 345

EP - 358

JO - Journal of Signal Processing Systems

JF - Journal of Signal Processing Systems

SN - 1939-8018

IS - 3

ER -