首页> 外文会议>Conference on Stereoscopic Displays and Virtual Reality Systems >Multiview autostereoscopic display of 36view using an ultra-high resolution LCD
【24h】

Multiview autostereoscopic display of 36view using an ultra-high resolution LCD

机译:使用超高分辨率LCD进行36View的多视图自动立体显示

获取原文

摘要

We have developed an autostereoscopic multi view display with 36view using 15.1" ultra-high resolution LCD. The resolution of LCD used for experiment is QUXGA of 3200x2400. RGB sub pixels are aligned as vertical lines and size of each sub pixel is 0.032 mm by 0.096mm. Parallax barrier are slanted at the angle of tan-1(1/6) = 9.46 degree and placed before LCD panel to generate viewing zones. Barrier patterns repeated approximately for every 6 pixels of LCD. So, the numbers of pixels decrease by six along the horizontal direction and the vertical direction. Nominal 3D resolution becomes (3200/6) x (2400/6) = 533 x 400. In slanted barrier configuration, the angular luminance profile for each zone overlaps each other. For the case of 2view 3D system, cross-talk between left eye and right eye zone deteriorates 3D image quality. However for multi view 3D, cross-talk between adjacent zones does not always bring about negative effects as image differences between adjacent zones are rather small. As viewers can see pixels from multiple viewing zones at one place, viewers feel 3D image of higher resolution than nominal 533x400 resolutions. Tested 3D images are made by computer graphics, in which camera position and depth of 3D objects are varied. Smooth motion parallax is observed for the limited depth range of 3D object. As depth of 3D object increases, 3D objects are observed not one image but as overlapped multiple images and image flipping becomes noticeable. We changed the barrier conditions so that horizontal angles between each zone are different and 3D image qualities were compared. For each barrier condition of different horizontal angle between viewing zones, we find an acceptable range of 3D object depth and camera displacement between each zone for computer generated images.
机译:我们开发了一种使用15.1“超高分辨率LCD的36View的自动立体多视图显示。用于实验的LCD的分辨率是3200x2400的Quxga。RGB子像素作为垂直线对齐,每个子像素的尺寸为0.032 mm 0.096毫米。视差屏障以TAN-1(1/6)= 9.46度的角度倾斜,并在LCD面板之前放置以产生观看区域。屏障模式重复大约每6个像素的LCD。因此,像素数量减少六沿水平方向和垂直方向。标称3D分辨率变为(3200/6)x(2400/6)= 533 x 400.在倾斜的屏障配置中,每个区域的角亮度曲线彼此重叠。对于2View 3D系统,左眼和右眼区域之间的串扰恶化了3D图像质量。然而,对于多视图3D,相邻区域之间的串扰并不总是带来负面影响,因为相邻区域之间的图像差异相当个小型湖由于观众可以在一个地方看到来自多个观看区域的像素,观众感觉到比标称533x400分辨率更高的分辨率的3D图像。测试的3D图像由计算机图形制成,其中相机位置和3D对象的深度变化。对于3D对象的有限深度范围,观察到平滑运动视差。随着3D对象的深度增加,观察到3D对象而不是一个图像,而是作为重叠的多个图像和图像翻转变得明显。我们改变了屏障条件,使每个区域之间的水平角度是不同的,并且比较了3D图像质量。对于观看区域之间的不同水平角度的每个障碍条件,我们在每个区域之间找到了可接受的3D对象深度和相机位移,用于计算机生成的图像。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号