[MUSIC] In the next few videos, we'll talk about applications in virtual reality. So far we've learned about VR hardware. We've learned that a typical high-end VR hardware would come with two parts. The first part is a VR display, which provides surrounding 3D stereo vision, and allows the user dynamic control of their view point, the precision and rotation tracking. The second part is a pair of VR controllers which support 3D user interaction with our hands. So just like how we use our hands in real life, in virtual reality, we can use our hands to select and manipulate virtual objects, and to gesture, and interact socially. Another way to look at VR and how it differs from other media is that it consists of three parts. The VR display, for instance, HMDs. VR interaction, which is usually supported by the VR controllers. And VR content, which refers to the images that are on display that you can sometimes interact with. Some people would argue that only real-time 3D graphics and animation count as VR content. But nowadays the term is very often used to include 360 degree video or images. We will discuss the differences between these two types of VR content later. Now I'd like to explain why it could be useful to divide VR into these three components, and how I use this as a framework to think about VR applications. Personally, I find this a useful framework for understanding VR applications. For instance, if I want to design a VR ping pong application, I start by thinking about how we play ping pong in real life, and how everything carries over to VR. I have to think that when you play ping pong or table tennis, it's important that you're able to move your body around. So to make it more natural, we will meet a VR display with position tracking. In this case, mobile VR would be less suitable. Secondly, you will need to hold your ping pong bat. Its position and orientation will have to be both tracked. And ideally when you hit the ball, you should be able to have some simple haptic feedback. Finally, for VR content we'll need both 3D graphics and animation. So we'll need to program a simple physical simulation to animate a ball using Newton's laws of motion. But the graphics and animation do not need to be very sophisticated to be effective, so there won't be any frame rate or latency concerns. So it's clear that hardware is available on the market to support this. We will need high-end VR hardware supported by desktops. There is a slight issue here, as those devices are currently not wireless. This is fine with a game like ping pong because you never really need to be particularly far from the table, so all you need is a relatively small space. But the same setup wouldn't work for tennis in which you have to move around a much bigger space. So ping pong will be a good VR application, as it is well supported by the current VR hardware. Let's have a look at another example. If you would like to design a VR shopping application which allows users to buy clothes, then you will realize that you might have some problems. Again, let's think about what you would do in real life. In real life when I shop for clothes, I would definitely want to try them out to see how they look on me. So VR display is not a problem here, as it can even be mobile VR. However, I would also like to be able to really feel the texture of the material. When it comes to VR interaction, the ability to actually feel the textures in VR would go beyond the current VR hardware. So this might not be a good idea. Another problem could be VR content. Ideally, you would have a big database of 3D models of clothes you are selling. But you would also need a physical model of the body of your customer. You could then use physical simulation to show how the clothes might fit on the body of your customer. The problem here is that 3D scanning and physical simulation of soft material are both very difficult challenges. You might say that you'd like to design an application where customers can just look at the clothes in a VR display without being able to touch them or try them on. In that case, my question is do you still need to be in VR? Why don't you just use a desktop 2D application instead? After all, VR is still more difficult to set up, and the resolution is still fairly low compared to an average desktop display. So with the current technology, shopping for clothes in VR might not add much to the online shopping applications, unless you have some clever ways to solve the problems mentioned previously. [MUSIC]
