Video compression dataset and benchmark of learning-based video-quality metrics

A. Antsiferova, S. Lavrushkin, M. Smirnov, A. Gushchin,
D. S. Vatolin, and D. Kulikov

Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS | 2022)

NeurIPS Poster | Benchmark Main Page | OpenReview Page | Dataset Page

Abstract

Video-quality measurement is a critical task in video processing. Nowadays, many implementations of new encoding standards - such as AV1, VVC, and LCEVC - use deep-learning-based decoding algorithms with perceptual metrics that serve as optimization objectives. But investigations of the performance of modern video- and image-quality metrics commonly employ videos compressed using older standards, such as AVC. In this paper, we present a new benchmark for video-quality metrics that evaluates video compression. It is based on a new dataset consisting of about 2,500 streams encoded using different standards, including AVC, HEVC, AV1, VP9, and VVC. Subjective scores were collected using crowdsourced pairwise comparisons. The list of evaluated metrics includes recent ones based on machine learning and neural networks. The results demonstrate that new no-reference metrics exhibit high correlation with subjective quality and approach the capability of top full-reference metrics.

Leaderboard

On our benchmark page you can find even more image and video quality metrics.