Super-multiview content with high angular resolution: 3D quality assessment on horizontal-parallax lightfield display

Abstract

The advent of glasses-free 3D super-multiview (SMV) displays has opened up new avenues for experiencing 3D content. Accordingly, perceptual quality assessment of such content assumes significance. In this context, we present the results of subjective and objective quality assessment of static 3D SMV content, conducted with twenty subjects on a glasses-free horizontal-parallax 3D lightfield display. First, we create a multiview image dataset using three real objects, with high angular resolution. The display system generates 3D views from fixed subsets of consecutive images. Next, we extend the standard guidelines of subjective assessment to our evaluation environment. In the course, we make recommendations on acquiring SMV data as well as conducting subjective evaluation. Subsequently, we observe that the existing 2D full-reference (FR) quality metrics, applied to individual 3D views, are inadequate for 3D quality assessment. To fill the gap, we propose a 3D FR objective quality metric. For every 3D view, the proposed metric combines spatial information from each constituent image and angular information (depth cues) from consecutive images. Finally, we show that the proposed metric correlates significantly with subjective scores, outperforming existing 2D metrics. The efficacy of pooling spatial and angular information highlights the fact that angular information plays a crucial role in 3D perception.