Image quality prediction using synthetic and natural codebooks: comparative results

TL;DR

This paper compares natural and synthetic image codebooks for quality assessment, proposing modifications to improve inference speed and accuracy, demonstrating real-time CPU performance and analyzing the impact of synthetic images on quality prediction.

Contribution

It introduces modifications to the codebook building method, evaluates synthetic images for codebook construction, and demonstrates potential for real-time quality assessment on CPU.

Findings

Synthetic images can enhance quality assessment accuracy.

Real-time CPU implementation is feasible with high correlation to MOS.

Different pooling strategies affect metric sensitivity to bitrate.

Abstract

We investigate a model for image/video quality assessment based on building a set of codevectors representing in a sense some basic properties of images, similar to well-known CORNIA model. We analyze the codebook building method and propose some modifications for it. Also the algorithm is investigated from the point of inference time reduction. Both natural and synthetic images are used for building codebooks and some analysis of synthetic images used for codebooks is provided. It is demonstrated the results on quality assessment may be improves with the use if synthetic images for codebook construction. We also demonstrate regimes of the algorithm in which real time execution on CPU is possible for sufficiently high correlations with mean opinion score (MOS). Various pooling strategies are considered as well as the problem of metric sensitivity to bitrate.