Performance Evaluation of EMOS Model for Mapping-based Video Quality Estimation

Accurate user-perceived video quality estimation models are increasingly needed with the proliferation of multimedia services. Previous research studies have focused on proposing and evaluating objective Video Quality Assessment (VQA) metrics, without mapping their values to Mean Opinion Scores (MOS). This paper presents a model to compute the estimated user-perceived video quality (EMOS), by combining multiple objective VQA metrics whose continuous values are mapped to discrete scores on the 0-5 MOS scale. The results analysis of a subjective video quality assessment study with 60 participants have shown that combining multiple VQA metric mappings can improve the user-perceived quality estimation accuracy up to 98.5%.