Stereoscopic video quality assessment model based on spatial-temporal structural information

Abstract

Most of the existing 3D video quality assessment methods estimate the quality of each view independently and then pool them into unique objective score. Besides, they seldom take the motion information of adjacent frames into consideration. In this paper, we propose an effective stereoscopic video quality assessment method which focuses on the inter-view correlation of spatial-temporal structural information extracted from adjacent frames. The metric jointly represents and evaluates two views. By selecting salient pixels to be processed and discarding the others, the processing speed is significantly improved. Experimental results on our stereoscopic video database show that the proposed algorithm correlates well with subjective scores.Index Terms-Stereoscopic video quality assessment, spatial-temporal structure, inter-view correlation, humanvisual system (HVS), asymmetric coding

Game theory based no-reference perceptual quality assessment for stereoscopic images

“…This subjective experiment has been published in [43]. Figure 6 and Table 2 show the details of the subjective test setting.…”

“…Figure 6 and Table 2 show the details of the subjective test setting. We choose several previously published 3D quality metrics to compare with our method, including PQM [43], PHVS-3D [44] and SFD [11].The performance comparison results are shown in Table 3. It is noted that the proposed method is NR quality assessment, others are RR or FR quality assessment algorithms.…”

Disparity weighting applied to full-reference and no reference stereoscopic image quality assessment

“…Nevertheless, this approach is limited because stereoscopic image signals have the disparity information, i.e., an inter-view relation, that must to be considered in the design of the objective algorithms [6], [7]. The challenges of the objective stereoscopic video quality assessment have encouraged the development of research in the subject [8]- [10].…”

ImQET

“…Recently, the disparity has been used as a depth estimation in the development of objective algorithms [3,4]. In addition, challenges in objective stereoscopic image quality assessment have encouraged the development of research in the subject [5,6]. Therefore, this paper presents a platform independent application with a GUI for objective stereoscopic IQA, called ImQET.…”