IEEE Transactions on Broadcasting, July 25, 2022
Why No Reference Metrics for Image and Video Quality Lack Accuracy and Reproducibility
doi: 10.1109/TBC.2022.3191059Abstract: This article provides a comprehensive overview of no reference (NR) metrics for image quality analysis (IQA) and video quality analysis (VQA). We examine 26 independent evaluations of NR metrics (previously published) and analyze 32 NR metrics on six IQA datasets and six VQA datasets (new results). Where NR metric developers claim Pearson correlation values between 0.66 and 0.99, our measurements range from 0.0 to 0.63. None of the NR metrics we analyzed are accurate enough to be deployed by industry. Performance evaluations that indicate otherwise are based on insufficient data and highly inaccurate. We will examine development strategies, tools, datasets, root cause analysis, and our baseline metric for collaboration, Sawatch.
Keywords: video quality; image quality; root cause analysis (RCA); no reference (NR); metric; Sawatch
For technical information concerning this report, contact:
Margaret H. Pinson
Institute for Telecommunication Sciences
(303) 497-3579
mpinson@ntia.doc.gov
To request a reprint of this report, contact:
Lilli Segre, Publications Officer
Institute for Telecommunication Sciences
(303) 497-3572
LSegre@ntia.gov
Disclaimer: Certain commercial equipment, components, and software may be identified in this report to specify adequately the technical aspects of the reported results. In no case does such identification imply recommendation or endorsement by the National Telecommunications and Information Administration, nor does it imply that the equipment or software identified is necessarily the best available for the particular application or uses.