Visual Quality Assessment of Panoramic Video

In contrast with traditional video, panoramic video enables spherical viewing direction with support for headmounted displays, providing an interactive and immersive experience. Unfortunately, to the best of our knowledge, there are few visual quality assessment (VQA) methods, either subjective or objective, for panoramic video. This paper proposes both subjective and objective methods for assessing quality loss in impaired panoramic video. Specifically, we first establish a new database, which includes the viewing direction data from several subjects watching panoramic video sequences. Then, from our database, we find a high consistency in viewing direction across different subjects. The viewing directions are normally distributed in the center of the front regions, but they sometimes fall into other regions, related to video content. Given this finding, we present a subjective VQA method for measuring different mean opinion score (DMOS) of the whole and regional panoramic video, in terms of overall DMOS (O-DMOS) and vectorized DMOS (VDMOS), respectively. Moreover, we propose two objective VQA methods for panoramic video, in light of human perception characteristics of panoramic video. One method weighs the distortion of pixels with regard to their distances to the center of front regions, which considers human preference in a panorama. The other method predicts viewing directions according to video content, and then the predicted viewing directions are leveraged to assign weights to the distortion of each pixel in our objective VQA method. Finally, our experimental results verify that both the subjective and objective methods proposed in this paper advance state-of-the-art VQA for panoramic video.