An Audio-Visual Quality Assessment Methodology in Virtual Reality Environment

PubDate: November 2018

Teams: Beijing Institute of Technology

Writers: Bo Zhang; Zhaoyu Yan; Jing Wang; Yiyu Luo; Shu Yang; Zesong Fei

PDF: An Audio-Visual Quality Assessment Methodology in Virtual Reality Environment

Abstract

Panoramic video and spatial audio are widely used in Virtual Reality (VR) experience. To evaluate the quality of experience with higher veracity and efficiency, this paper presents a quality assessment methodology for audio-visual multimedia in virtual reality environment. By employing the audiovisual sequences with multiple combinations of video, audio and their alignment distortions, we conduct subjective assessment experiment and obtain user perception. The quality factors which represent different distortions are applied as the input to neural network. And thus the audio-visual objective quality assessment model is established. Ulteriorly, the performance of the model is evaluated, proving that the proposed methodology is suitable for audio-visual quality assessment in VR environment.