Questions & explanations
1. What technical challenges make real-time VR streaming hard?
The main challenge is latency: the time between a movement and seeing the result must be very short (under 20 milliseconds) to avoid motion sickness. Sending high-resolution 3D video over the internet takes a lot of data. The connection must be fast and stable. Compression and decompression add delay. Also, the viewpoint changes as the user moves, so the system must update the stream instantly. Another challenge is synchronizing audio and video from different people. To solve these, engineers use special algorithms and sometimes run the rendering locally with only updates from the remote site. Even with 5G, it is still difficult to achieve perfect quality.
2. Can immersive art be considered 'political'? Give a theoretical perspective.
Yes, any art can be political, and immersive art is no exception. Theory says that even if the artist does not intend it, the technology and choices carry political meanings. For example, a VR simulation of a refugee camp might claim to build empathy, but critics might say it reduces real suffering to a spectacle. The act of putting on a headset can create a distance: you can leave the experience anytime, while real refugees cannot. Another angle: the data collected in VR (your movements, gaze) raises privacy issues. So immersion is not neutral; it involves power, representation, and ethics. The critique pushes artists to be aware of these layers.
3. How does sensory integration affect presence in VR?
Sensory integration means combining what you see, hear, and feel into one experience. In VR, if your visual movement matches the motion your inner ear feels (vestibular system), you feel present. But if there is a mismatch, like seeing movement while sitting still, you might feel sick. Adding touch (haptics) can boost presence: if you see a virtual object and feel a vibration in your hand, your brain believes it is real. Sound also helps: footsteps that match your walking make you feel more grounded. Good VR integrates all senses smoothly. When senses agree, the illusion of being in the virtual world is strong. Disagreement breaks the feeling.
4. What is telepresence in the context of VR installations?
Telepresence means that a person in one location can feel as if they are present in a remote location through VR. They wear a headset that shows a live 3D video or reconstructed scene from the other place. They can look around and sometimes move, just like being there. The remote location may have cameras and sensors that stream the data in real time. The person can also interact with people or objects at the remote site using their hands or controllers. This allows someone to attend a meeting, visit a museum, or explore a dangerous place without traveling. The goal is to make the remote experience feel as real as being there physically.
5. How does proxemics change when people interact in VR?
Proxemics is the study of how close people stand to each other in real life. In VR, people often stand closer to avatars than to real strangers because they know it is not real. But if the avatar looks very human, some people keep the same social distance as in real life. The virtual space also changes things: you can teleport or float, so distance is less important. However, if a virtual character invades your personal space (comes too close), you might still feel uncomfortable. Experiments show that your brain treats virtual personal space similarly to real space. So VR designers must respect this to avoid making users feel crowded.
6. How do people interact with each other in a networked VR installation?
They can see each other's avatars, which move as they move. They can speak through microphones and hear each other's voices as if from the avatar's location. They can point at things, wave, and use hand gestures if their controllers track fingers. Some setups allow real hand tracking using cameras. They can also touch and manipulate virtual objects together: for example, both can grab the same virtual ball and throw it. The system must handle collisions and ownership of objects. The goal is to make social interaction feel natural despite being in different places. This allows collaboration on tasks like building or playing games.
7. What is the difference between 'immersion' and 'presence' according to theory?
Immersion is the objective quality of the technology: it has a wide field of view, good sound, and tracks your movements. Presence is the subjective feeling of 'being there'. You can have high immersion but still feel like you are in a simulation if the content is boring. Conversely, a simple 2D screen can give strong presence if the story draws you in. In theory, presence is a mental state, while immersion is a property of the system. Critics argue that artists often confuse the two, thinking better graphics mean better art. Understanding this difference helps evaluate whether a work truly engages you or just uses flashy tech.
8. What is the difference between streaming a 360 video and a full 3D VR scene for telepresence?
A 360 video is a flat video wrapped around you; you can look around but cannot move your head position. A full 3D VR scene, like a NeRF or Gaussian splatting, lets you move your whole body and see objects from different sides. 360 video is easier to stream because it is just a video, but you feel stuck in one spot. 3D telepresence feels more real because you can step closer to objects. However, it needs much more data and processing. For many applications, 360 video is good enough for simple virtual visits. But for true presence, 3D is better. Some systems combine both: a 3D model for the room and 360 video for the background.
9. Give an example of a VR experiment that studies body ownership.
In a classic experiment, people wear a VR headset and see a virtual body from a first-person view. A virtual ball approaches and 'hits' the virtual stomach, and the person flinches as if hit in reality. Another experiment: people see a virtual mirror showing an avatar that looks like them but with a different skin color. They move, and the avatar moves identically. After a while, they feel that avatar is their own body. Then the avatar's skin color changes, and researchers measure changes in racial bias. This shows that body ownership can change attitudes. These studies help understand how our brain builds a sense of self.
10. What is a 'body illusion' in VR?
A body illusion is when you feel that a virtual body or part of a body belongs to you. For example, you see a virtual hand that moves when you move your real hand, and soon you feel the virtual hand is your own. The rubber hand illusion is a famous real-world experiment, and VR can create stronger versions. You can even feel a whole virtual body as yours. This happens because your brain combines what you see with what you feel. If a virtual arm is touched and you see it, you might feel the touch on your real arm. These illusions show how flexible our sense of self is. They are used in psychology and art to study identity.
11. What is the 'rubber hand illusion' and how is it done in VR?
In the real rubber hand illusion, a fake hand is placed on a table while your real hand is hidden. Both hands are stroked at the same time, and you start to feel the fake hand is yours. In VR, you see a virtual hand that matches your real hand's movements. The virtual hand can be stroked with a virtual brush while your real hand is stroked in sync. The illusion works even if the virtual hand looks different, like a cartoon or a claw. The key is timing: seeing and feeling touch at the same moment. VR makes it easy to change the hand's shape or color, allowing researchers to test what limits our sense of body ownership.
12. Compare NeRF and Gaussian splatting in terms of memory use.
A NeRF model is a neural network with a fixed number of parameters, usually a few megabytes. It stores the scene compactly but takes more computation to render. Gaussian splatting stores thousands to millions of individual Gaussians, which can be larger in memory (tens to hundreds of megabytes). However, the memory use grows with the scene complexity: a simple scene uses fewer Gaussians. NeRF's memory is the same regardless of scene size because the network size is fixed. For very large scenes, NeRF may be more memory-efficient. But Gaussian splatting's memory can be controlled by adjusting the number of Gaussians.