虚拟现实
计算机科学
失真(音乐)
色差
色阶
计算机图形学(图像)
计算机视觉
人工智能
光学
物理
电信
放大器
带宽(计算)
作者
Yining Li,Congshan Rui,Le Zhang,Mengna Zhao,Chaohao Wang,Lei Zhao
摘要
At present, a majority of virtual reality (VR) technologies on the market employ static distortion correction by predistorting the virtual image. However, this compensation method is only effective when the pupil remains in a fixed position for virtual display device. When the pupil moves within the eye box of the VR device, the virtual image may deviate from the target position, rendering the compensation ineffective. Due to the optical asymmetry of the lens, different distortions can be perceived by the human eye as the pupil moves, which adversely affects the user's visual experience. Therefore, it is essential to measure and evaluate dynamic distortion for adjusting pre-compensation parameters according to the pupil's position, as well as for further optimizing optical systems with low dynamic distortion. In this paper, we analyzed the cause of dynamic distortion in virtual reality and proposed a novel method for characterizing dynamic distortion, allowing for quantitative analysis of dynamic distortion compared to traditional optical flow maps. A prototype was fabricated for dynamic distortion evaluation, and both simulation and measurement of the dynamic distortion were conducted. The results demonstrate a strong correlation between the simulations and measurements.
科研通智能强力驱动
Strongly Powered by AbleSci AI