IEEE Std 3333.1.1-2015 pdf free download – IEEE Standard for Quality of Experience (QoE) and Visual-Comfort Assessments of Three-Dimensional (3D) Contents Based on Psychophysical Studies.
4. Ergonomic requirements and recommendations
4.1 General Unlike in two-dimensional (2D) video, the ocular adjustment to 3D depth can induce neurological symptoms, such as visual discomfort and headache, and 3D distortions that cause quality degradation. Understanding these problems involves several intricate visual factors that can only be probed by investigating the reliable spatial and temporal features in 3D contents and by using a reliable subjective testing methodology.
4.2 Visual saliency prediction When viewing a 3D image and video, some scenes have only a few regions that cause visual discomfort and low QoE states. In these cases, the viewer tends to pay more attention to this region, and the rest of the image is projected onto the retina with a low density of photoreceptors. Therefore, to develop reliable human factor models and a quality assessment metric, a concrete visual saliency algorithm shall be developed.
4.3 Visual contents analysis To analyze and predict the degree of the QoE or visual comfort when viewing 3D content, it is necessary to understand the contents in terms of spatial and temporal characteristics that are based on existing psychophysical and statistical models of 3D visual perception.
4.4 Subjective assessment Most subjective assessment is inherited from what has been traditionally done for 2D subjective assessment as defined by ITU-T P.910, ITU-R BT.500-13, and ITU-R BT.2021. However, it is doubtful whether these results are reliable enough to be used as references because the viewing environment is quite different from 2D due to the intensive immersion of a user wearing the glasses in a the dark. Hence, to perform the subjective 3D image and video assessments, a novel interface shall be designed that covers the characteristics of the human perception, display mechanism, viewing environment, and so on.
5. Visual saliency prediction
5.1 General An important ingredient in further improving 3D video processing technologies is the effort to incorporate better models of 3D perception. Among these, saliency detection, or the automated discovery of points of high visual interest, conspicuity, or task relevance, is a challenging problem. This clause describes visual saliency prediction method by considering human visual system (HVS) characteristics.
5.2 Human visual system To capture human factors or predict the visual discomfort of 3D contents, the characteristics of the HVS such as a retina and fovea shall be considered. The fovea is responsible for sharp central vision, which is needed in human beings for reading, watching television or movies, driving, and any activity for which visual detail is of primary importance. As shown in Figure 1, photo-receptors possess non-uniform spatial distribution with the highest density at the fovea, and the density decreases dramatically with distance from the fovea. Here we describe how the region with the highest saliency (obtained as described above) falls on the fovea and, hence, has the highest sensitivity/resolution.
5.3 Saliency prediction
5.3.1 2D saliency features
This standard considers four 2D visual saliency features: luminance, color, size, and compactness.The luminance contrast and gradient of stereoscopically fixated patches are generally higher than in randomly selected patches. Thus these features shall be represented as.IEEE Std 3333.1.1 pdf download.