(A) quality perception assessment model for stereoscopic video using disparity information스테레오스코픽 비디오의 시차(視差) 정보를 이용한 품질 인지 모델

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 578
  • Download : 0
According to increasing demand for realistic broadcasting which includes presence and realism, 3-Dimensional (3D) video and Ultra High-Definition (UHD) video has attracted considerable attention recently beyond the existing High-Definition (HD) video. Specially, 3DTV is expected as a promising next-generation broadcasting to maximize sensation of presence by providing an additional dimension, depth. Besides, visual discomfort/fatigue problems such as vomiting, sickness, dizziness caused by viewing 3D video, have become more important issues. Because 3DTV displays the same image shot from slightly different angles to each eye artificially in order to create the illusion of 3-D scenes. So safety problems on stereoscopic video contents may occur more frequently. Therefore, 3D video quality and visual comfort evaluation tools are essential to prevent from the above problems and to guarantee the safety of 3D contents. In this study, we examine the factors affecting human perception of depth and visual comfort from stereoscopic video which has little 2D artifacts such as blockiness, blurring. For this, we conduct subjective quality assessments using the Single Stimulus (SS) method of ITU-R BT.500-11 on 105 stereoscopic test video sequences of HD resolution for twenty subjects. Here the subjective quality assessments are evaluated in terms of visual quality, perceived depth and visual comfort. After analyzing these results, we extract four factors: (1) Mean of motion vector`s magnitudes is extracted to represent temporal variance; (2) Disparity variation in intra-frames and (3) disparity variation in inter-frames are used because excessive change of depth can cause visual fatigue; (4) disparity distribution of frame boundary areas is used. Finally we combine these four factors to propose a no-reference stereoscopic video quality perception model (SV-QPM). The model parameters are estimated using a linear regression model based on the results of subjective quality a...
Advisors
Kim, Mun-Churlresearcher김문철researcher
Description
한국과학기술원 : 전기 및 전자공학과,
Publisher
한국과학기술원
Issue Date
2011
Identifier
467797/325007  / 020094097
Language
eng
Description

학위논문(석사) - 한국과학기술원 : 전기 및 전자공학과, 2011.2, [ iv, 41 p. ]

Keywords

stereoscopic video quality assessment; no-reference; quality perception model; 스테레오스코픽 품질 평가; 무기준법; 품질 인지 모델; MOS 예측; mean opinion score prediction

URI
http://hdl.handle.net/10203/180819
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=467797&flag=dissertation
Appears in Collection
EE-Theses_Master(석사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0