3D video quality assessment based on visual perception

Abstract

This paper proposes a new 3D video quality assessment based on 3D visual perception for texture and depth image for measuring the quality of stereoscopic 3D videos by detecting 3D distortions. The effectiveness of the proposed metrics is verified by conducting subjective evaluations on publicly available stereoscopic image databases. Experimental results show that significant consistency can be reached between the measured MOS and the proposed metrics, indicating that the proposed metric in comparison to 2D Quality metrics and other reference metrics is more suitable to human perception, as demonstrated by Pearson correlation coefficient (PLCC) and SRCC obtained from subjective and objective scores. Because the proposed 3D video quality assessment is based 3D visual perception, the quality assessment results bear more resemblance to the 3D experience of the viewer.