Video quality assessment (VQA) is the academic name for the field that studies how to quantify perceived video quality, spanning both subjective methods, where humans rate clips, and objective methods, where an algorithm computes a score. In practice the two are tightly linked: subjective testing produces the ground-truth Mean Opinion Score, and objective metrics are built and graded by how closely their output tracks that score, using statistics such as the Pearson and Spearman correlations and RMSE defined in ITU-T P.1401. The catch is that the term covers a broad research program, not a single tool, and quality is multidimensional, so no one method answers every case. VQA is the umbrella under which the reference setups, the objective metrics PSNR, SSIM, and VMAF, and the subjective standards P.910 and BT.500 all sit, and it is essentially synonymous with the engineering term video quality measurement.
The handbook · 7 volumes
Fora Soft Video Engineering Handbook
Video quality assessment (VQA) covers both people rating clips and algorithms predicting those ratings, and products need both. Our seven-volume handbook spans the whole video stack; for VQA start with Volume 1, covering subjective testing, full- and no-reference quality metrics, VMAF, SSIM and PSNR, then Volume 3 for QoE in delivery. Written by Nikolay Sapunov, CEO at Fora Soft.
Setting up video quality assessment for your product from scratch?
We will design a video quality assessment plan that combines metrics, subjective checks and QoE data for your pipeline. 250+ video products shipped since 2005.
