Comparison of Differences between Human Eye Imaging and HMD Imaging

As virtual reality (VR) headsets with head-mounted displays (HMD) have attracted more and more attention, new research questions have emerged. In virtual reality games, virtual reality movies or virtual reality simulations, maintaining a high degree of depth perception and minimizing visual discomfort is essential to improve the overall user experience. Proprioceptive cues, monocular cues, and binocular cues constitute the depth cues of visual perception, which accurately provide the depth information of objects in personal space. Proprioceptive cues include convergence and adjustment. Quite a few studies have proved that convergence and accommodation in VE exacerbate uncomfortable feelings. Monocular cues include occlusion, relative size of objects, texture fineness, variations of light and the shadows, optical thickness, perspective, etc. Monocular clues produce psychological depth cues, which can be affected by the texture mapping of objects and environment settings in the virtual environment. The binocular vision of the human eye enables users to perceive the depth of different objects in space. When both eyes observe the same target object in reality, the images formed on the retina are different due to the different lateral positions of the two eyes. The stimulus received by the retina is processed in the brain to generate stereo vision. The perception of objects generated by binocular vision is influenced by depth cues in environment. The difference in depth perception of virtual information leads to the separation of spatial perception, which makes it difficult to establish relevance with the reality operation mode. In this paper, the physiological characteristics of human eyes are studied, and visual environment provided by head-mounted displays (HMD) equipment are summarized. The limitations of visual images provided by HMD are obtained by comparing the differences. We discussed imaging mechanism of the stereo vision in HMD. By comparing the differences of stereo vision generated in HMD images and human eyes, this paper is aimed to provide theoretical suggestions for building a more realistic and immersive environment in VR.