A Human Visual System-Based 3D Video Quality Metric

TL;DR

This paper introduces a novel full-reference 3D video quality metric based on the human visual system, effectively correlating with subjective quality assessments and considering multiple visual factors.

Contribution

It presents a new 3D quality metric that integrates binocular vision, color, and disparity, advancing beyond existing 2D metrics for 3D content evaluation.

Findings

Achieves 86% correlation with subjective quality scores

Successfully detects various types of 3D distortions

Incorporates cyclopean view and disparity information

Abstract

Although several 2D quality metrics have been proposed for images and videos, in the case of 3D efforts are only at the initial stages. In this paper, we propose a new full-reference quality metric for 3D content. Our method is modeled around the HVS, fusing the information of both left and right channels, considering color components, the cyclopean views of the two videos and disparity. Performance evaluations showed that our 3D quality metric successfully monitors the degradation of quality caused by several representative types of distortion and it has 86% correlation with the results of subjective evaluations.