Abstract
3D displays have become the mainstream due to the development of display manufacturers. However, there have been some problems existed in current display technology. The crosstalk issue stemming from the leakage of light between viewing zones corresponding to the left and the right eyes is inevitable in some autostereoscopic displays and could sometimes be serious. 3D cues are also the major factors affecting viewers’ visual quality. These phenomena of 3D displays may increase viewers’ mental and physical loading. 3D TVs and movies have entered the mass market; however, the inadequate display luminance and ambient illumination might affect 3D image quality for viewers. As a result, it is a critical issue how to enhance the image quality of 3D images and the subject’s visual comfort on viewing 3D displays. This study is composed of two experiments. The first research studied the effect of crosstalk and 3D cues upon viewers’ visual fatigue and image quality evaluation in both mirror-type 3D displays and barrier-type autostereoscopic displays. The first ergonomic experiment included three parts. First, the effect of system crosstalk, shadow, and linear perspective of 3D pictures in a mirror-type 3D display was verified. Second, an acceptable threshold level of system crosstalk was analyzed. Third, the fitting system crosstalk level in barrier-type autostereoscopic displays corresponding to that in mirror-type 3D displays was compared. The results showed that system crosstalk, shadow and linear perspective significantly affected image quality evaluation. The acceptable system crosstalk level in mirror-type 3D displays ranged from 18.18% to 22.98%. The fitting system crosstalk level in a barrier-type autostereoscopic display ranged from 21.52% to 30.46%, which was higher than that in a mirror-type 3D. In addition, the combination of 3D cues, grid and linear perspective, in a content of 3D images shown in a barrier-type autostereoscopic display could enhance the image quality evaluation. However, viewers replied neither visual fatigue nor visual discomfort when watching 3D pictures for only 10 minutes in this experiment. The second experiment extended the first experiment to investigate whether screen luminance and ambient illumination have significant effects on the perception of 3D television imagery for static pictures and dynamic films. Two kinds of stimuli were shown on the multi-view stereoscopic display: the static pictures, which included computer-generated and photographic images, and the dynamic films, which contained real-life and animation images. In each treatment with a different level of screen luminance, subjects finished psychophysical and physiological measurements and subjective comfort evaluation. The results showed the suitable watching condition of luminance and ambient for 3D viewing experience. When subjects viewed 3D static pictures, ambient illumination of 55 Lux could produce the lower psychophysical visual fatigue, and screen luminance of 34.2 cd/m2 could cause better subjective comfort evaluation and visual discrimination performance. However, when subjects watched 3D dynamic films, screen luminance of 60 cd/m2 under brighter (300 Lux) or dark (0 Lux) ambient illumination was the better viewing condition.The results of this study could be served as a reference for the design of 3D displays. In order to promote the 3D display in the market, industry will need to tackle all of these challenges to support content creators and provide features and ease of use that end users require. For the further study, it is expected to obtain the optimal 3D cues, and the various impactful viewing 3D factors of dynamic films to build up guidelines for a qualified autostereoscopic and stereoscopic display.