Advanced Metrics for Quality of Experience in Multimedia Systems

Supervisor

Prof. T. Ebrahimi

Description

Measurement of perceived quality plays a fundamental role in the context of multimedia services and applications. Subjective quality evaluation is time-consuming and it cannot be applied when real-time in-service quality evaluation is needed. The research in objective quality assessment of visual and audio signals thus aims at developing quantitative measures, i.e. metrics, that can automatically predict perceived quality of the visual and audio content.

In this thesis, starting from the investigation of existing approaches for the objective quality assessment of the audio and visual contents, we focus on the No-Reference scenario, in which we have access only to the distorted signal, while the original (reference) signal is not available.

We aim at developing a new approach of NR objective quality assessment, which considers the integration of high-level features of the human quality perception, e.g. the Focus Of Attention (FOA) feature, according to the context of the considered application, which defines the multimedia experience involving the subject’s perception. We first consider the visual and audio contexts as separate areas of interest, but we also believe that the perception of the visual signal and the perception of the audio signal strictly influence each other and are both application-context-dependent. Therefore, an important part of our research concentrates on the understanding and the modeling of the multimodal perception of quality, in order to design a metric for the assessment of the more general and complex concept of Quality of Experience in a multimedia service.

A list of the papers published so far can be found here.