VR系列——Oculus最佳实践：三、视野和比例

虚拟镜头的视野（FOV）必须与可见的显示区域相匹配。一般情况下，Oculus公司建议不要改变默认视野。

视野可以让我们从一开始就区分不同事物。如果用术语“显示视野”（dFOV）来表示，我们指的是用户实际视野观察到的VR内容部分。这是一个物理硬件和光学的特性。视野还有一种类型是镜头视野（cFOV），指的是在任何时刻渲染镜头捕捉到的虚拟世界的范围。所有的视野都是通过垂直、水平，和（或）对角线的角度测量来定义的。

在普通的基于屏幕的计算机图形中，你通常能自由设置你想要的镜头视野：从鱼眼镜头（广角）一直到长焦镜头（窄角）。尽管人们会在游戏画面中遇到一些虚拟幻境头晕，[1]但通常不会影响到大部分用户，因为这些画面只是观察者对于周围环境总视野里的一部分。一个计算机用户的外围视野可以看见他们目前所处的房间，并且显示器通常不会影响用户的头部动作。即使当这些画面增加了沉浸感时，大脑也通常不会傻到以为它是真实的，cFOV和dFOV之间的差异对大部分人来说不会产生问题。

在VR中，这里没有外部空间视野，虚拟世界充满了你大部分的周边视野。因此cFOV和dFOV精准的匹配是非常重要的。这两个值的比值称为scale（比例），而且在VR中这比例必须刚好是1.0。

在Rift中，dFOV的最大值是由屏幕，镜头，和用户所调整的镜头与眼睛的距离（一般来说，眼睛越靠近镜头，dFOV越广）来决定的。配置设备测量出用户可以看见的dFOV的最大值，把这些信息储存在他们的个人资料中。然后SDK会基于这些信息，推荐一个cFOV来匹配dFOV。

注意：一些人的其中一个眼睛会更靠近屏幕，所以每个眼睛会有不同的dFOV。这是正常的。

dFOV和cFOV之间的偏差令人感到不适 [2]（虽然一些研究主题不同[3]）。如果比例偏离了1.0，失真校正的值将导致已渲染的画面弯曲。操纵镜头视野也会引起虚拟幻境头晕，甚至导致前庭眼动反射适应不良，一种在头部运动时使眼睛在对象上保持稳定的反射机制。这些不良适应不仅会使用户在整个VR体验中感觉难受，而且在摘除Rift后也会影响视觉运动功能。

在不改变比例的前提下，SDK可以通过在画面周边补全黑边来调整cFOV和dFOV。使用较小的可视画面可以帮助提高渲染性能或者特效；请注意，如果你选择一个40°的可视画面，那么大部分的屏幕将是黑色的——这完全是有意的，并不是一个bug。还要注意的是如果可视画面较大，在调小画面时需要用户更多的头部运动来环顾四周；这些会导致肌肉疲劳和虚拟幻境头晕。

一些游戏存在缩放模式，如双筒望远镜或狙击。这在VR中是非常棘手的，必须更谨慎的去处理，如果只是简单的实现缩放，不仅将导致头部动作和视光学运动间对虚拟世界感知的差异，而且会引起不适。在这可以找到未来的博客和演示：

[1]Stoffregen, T.A., Faugloire, E., Yoshida, K., Flanagan, M.B., &Merhi, O. (2008). Motion sickness and postural sway in console video games（《在电子游戏中的晕动病和姿势晃动》）. Human Factors, 50, 322-331.

[2]Draper, M.H.
4000
, Viire, E.S., Furness, T.A., Gawron, V.J. (2001). Effects of image scale and system time delay onsimulator sickness with head-coupled virtual environments（《图像比例尺和系统时间延迟对虚拟幻境头晕的影响》）. Human Factors, 43(1), 129-146.

[3]Moss, J. D., &Muth, E. R. (2011). Characteristics of Head-Mounted Displays and Their Effects on SimulatorSickness（《头盔式显示器的特征和它们对虚拟幻境头晕的影响》）. Human Factors: The Journal of the Human Factors and Ergonomics Society, 53(3), 308–319.

原文如下

The FOV of the virtual cameras must match the visible display area. In general, Oculus recommends not changing with the default FOV.

Field of view can refer to different things that we will first disambiguate. If we use the term display field of view (dFOV), we are referring to the part of the user’s physical visual field occupied by VR content. It is a physical characteristic of the hardware and optics. The other type of FOV is camera field of view (cFOV), which refers to the range of the virtual world that is seen by the rendering cameras at any given moment. All FOVs are defined by an angular measurement of vertical, horizontal, and/or diagonal dimensions.

In ordinary screen-based computer graphics, you usually have the freedom to set the camera’s cFOV to anything you want: from fisheye (wide angle) all the way to telephoto (narrow angle). Although people can experience some visually-induced motion sickness from a game on a screen,[1] this typically has little effect on many users because the image is limited to an object inside the observer’s total view of the environment. A computer user’s peripheral vision can see the room that their display sits in, and the monitor typically does not respond to the user’s head movements. While the image may be immersive, the brain is not usually fooled into thinking it is actually real, and differences between cFOV and dFOV do not cause problems for the majority of people.

In virtual reality, there is no view of the external room, and the virtual world fills much of your peripheral vision. It is therefore very important that the cFOV and the dFOV match exactly. The ratio between these two values is referred to as the scale, and in virtual reality the scale should always be exactly 1.0.

In the Rift, the maximum dFOV is determined by the screen, the lenses, and how close the user puts the lenses to their eyes (in general, the closer the eyes are to the lens, the wider the dFOV). The configuration utility measures the maximum dFOV that users can see, and this information is stored inside their profile. The SDK will recommend a cFOV that matches the dFOV based on this information.

Note: Because some people have one eye closer to the screen than the other, each eye can have a

different dFOV. This is normal.

Deviations between dFOV and cFOV have been found to be discomforting[2] (though some research on this topic has been mixed[3]). If scale deviates from 1.0, the distortion correction values will cause the rendered scene to warp. Manipulating the camera FOV can also induce simulator sickness and can even lead to a maladaptation in the vestibular-ocular reflex, which allows the eyes to maintain stable fixation on an object during head movements. The maladaptation can make the user feel uncomfortable during the VR experience, as well as impact visual-motor functioning after removing the Rift.

The SDK will allow manipulation of the cFOV and dFOV without changing the scale, and it does so by adding black borders around the visible image. Using a smaller visible image can help increase rendering performance or serve special effects; just be aware that if you select a 40° visible image, most of the screen will be black—that is entirely intentional and not a bug. Also note that reducing the size of the visible image will require users to look around using head movements more than they would if the visible image were larger; this can lead to muscle fatigue and simulator sickness.

Some games require a “zoom” mode for binoculars or sniper scopes. This is extremely tricky in VR, and must be done with a lot of caution, as a naive implementation of zoom causes disparity between head motion and apparent optical motion of the world, and can cause a lot of discomfort. Look for future blog posts and demos on this.

[1] Stoffregen, T.A., Faugloire, E., Yoshida, K., Flanagan, M.B., & Merhi, O. (2008). Motion sickness and postural sway in console video games. Human Factors, 50, 322-331.

[2] Draper, M.H., Viire, E.S., Furness, T.A., Gawron, V.J. (2001). Effects of image scale and system time delay on simulator sickness with head-coupled virtual environments. Human Factors, 43(1), 129-146.

[3] Moss, J. D., & Muth, E. R. (2011). Characteristics of Head-Mounted Displays and Their Effects on Simulator Sickness. Human Factors: The Journal of the Human Factors and Ergonomics Society, 53(3), 308–319.