Enhancing VMAF through New Feature Integration and Model Combination

TL;DR

This paper improves the VMAF video quality assessment method by integrating new features and combining models, resulting in higher correlation with subjective opinions across diverse video databases.

Contribution

It introduces new video features and model combination techniques to enhance VMAF's accuracy and robustness across varied content and distortions.

Findings

Enhanced VMAF outperforms original VMAF (0.6.1)

Achieves higher correlation with subjective data

Effective across multiple HD video databases

Abstract

VMAF is a machine learning based video quality assessment method, originally designed for streaming applications, which combines multiple quality metrics and video features through SVM regression. It offers higher correlation with subjective opinions compared to many conventional quality assessment methods. In this paper we propose enhancements to VMAF through the integration of new video features and alternative quality metrics (selected from a diverse pool) alongside multiple model combination. The proposed combination approach enables training on multiple databases with varying content and distortion characteristics. Our enhanced VMAF method has been evaluated on eight HD video databases, and consistently outperforms the original VMAF model (0.6.1) and other benchmark quality metrics, exhibiting higher correlation with subjective ground truth data.