Loughborough University
Leicestershire, UK
LE11 3TU
+44 (0)1509 263171
Loughborough University

Loughborough University Institutional Repository

Please use this identifier to cite or link to this item: https://dspace.lboro.ac.uk/2134/20598

Title: Stereoscopic image quality assessment method based on binocular combination saliency model
Authors: Liu, Yun
Yang, Jiachen
Meng, Qinggang
Lu, Zhihan
Song, Zhanjie
Gao, Zhiqun
Keywords: Binocular vision
Visual attention
Image quality
Human visual system
Issue Date: 2016
Publisher: © Elsevier
Citation: LIU, Y. ... et al., 2016. Stereoscopic image quality assessment method based on binocular combination saliency model. Signal Processing, 125, pp.237-248.
Abstract: The objective quality assessment of stereoscopic images plays an important role in three-dimensional (3D) technologies. In this paper, we propose an effective method to evaluate the quality of stereoscopic images that are afflicted by symmetric distortions. The major technical contribution of this paper is that the binocular combination behaviours and human 3D visual saliency characteristics are both considered. In particular, a new 3D saliency map is developed, which not only greatly reduces the computational complexity by avoiding calculation of the depth information, but also assigns appropriate weights to the image contents. Experimental results indicate that the proposed metric not only significantly outperforms conventional 2D quality metrics, but also achieves higher performance than the existing 3D quality assessment models.
Sponsor: The authors would like to thank Prof. Alan C. Bovik for providing the LIVE 3D IQA Database. This research is partially supported by the National Natural Science Foundation of China (Nos. 61471260 and 61271324), and Program for New Century Excellent Talents in University (NCET-12-0400).
Version: Accepted for publication
DOI: 10.1016/j.sigpro.2016.01.019
URI: https://dspace.lboro.ac.uk/2134/20598
Publisher Link: http://dx.doi.org/10.1016/j.sigpro.2016.01.019
ISSN: 0165-1684
Appears in Collections:Closed Access (Computer Science)

Files associated with this item:

File Description SizeFormat
paper20150307.pdfAccepted version2.55 MBAdobe PDFView/Open

 

SFX Query

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.