PROJECT TITLE :

Blind Stereoscopic Video Quality Assessment From Depth Perception to Overall Experience - 2018

ABSTRACT:

Stereoscopic video quality assessment (SVQA) may be a challenging drawback. It has not been well investigated on how to live depth perception quality independently under totally different distortion classes and degrees, particularly exploit the depth perception to assist the overall quality assessment of 3D videos. During this Project, we propose a new depth perception quality metric (DPQM) and verify that it outperforms existing metrics on our revealed 3D video extension of High Potency Video Coding (3D-HEVC) video database. Furthermore, we validate its effectiveness by applying the crucial part of the DPQM to a unique blind stereoscopic video quality evaluator (BSVQE) for overall 3D video quality assessment. Within the DPQM, we have a tendency to introduce the feature of auto-regressive prediction-primarily based disparity entropy (ARDE) measurement and therefore the feature of energy weighted video content measurement, that are inspired by the free-energy principle and therefore the binocular vision mechanism. Within the BSVQE, the binocular summation and distinction operations are integrated along with the fusion natural scene statistic measurement and therefore the ARDE measurement to reveal the key influence from texture and disparity. Experimental results on 3 stereoscopic video databases demonstrate that our technique outperforms state-of-theart SVQA algorithms for each symmetrically and asymmetrically distorted stereoscopic video pairs of various distortion types.