The problem of stereoscopic image quality assessment, which findsapplications in 3D visual content delivery such as 3DTV, is investigated inthis work. Specifically, we propose a new ParaBoost (parallel-boosting)stereoscopic image quality assessment (PBSIQA) system. The system consists oftwo stages. In the first stage, various distortions are classified into a fewtypes, and individual quality scorers targeting at a specific distortion typeare developed. These scorers offer complementary performance in face of adatabase consisting of heterogeneous distortion types. In the second stage,scores from multiple quality scorers are fused to achieve the best overallperformance, where the fuser is designed based on the parallel boosting ideaborrowed from machine learning. Extensive experimental results are conducted tocompare the performance of the proposed PBSIQA system with those of existingstereo image quality assessment (SIQA) metrics. The developed quality metriccan serve as an objective function to optimize the performance of a 3D contentdelivery system.
展开▼
