I think this is where VR will eventually settle in for the majority of people. In a way, smartphones gave us the ability to communicate with anyone in the world, for practically free, from anywhere in the world. With no barrier, we just started doing it. VR is going to do the same thing for face-to-face communication.
Sure, there is Skype and Face Time and whatnot, but they aren't the same. You get stuck looking back and forth between the camera and the screen. The camera angles are weird and the receiving user is at the mercy of the person they're talking to as to what they're going to look at. Ironically, I've felt more personally engaged, more like I'm talking to a real person, having conversations in VR with robot-shaped avatars than I've ever had with human faces in Skype.
You don't run video over the VR session, it's all a 3D rendering. You have an avatar, and you correlate the HMD head movements to the avatar's 3D model. Combined with simple hand tracking like Leap Motion, you get very natural body language of of it. Check out AltspaceVR for an example.
There are also some fairly recent advances in photogammetry that I think we will soon all have 1-to-1 avatars with our own faces.
Note that in the future VR googles will have eye tracking to improve the efficiency of the rendering, so you could also use this data to render correctly the face.
Sure, there is Skype and Face Time and whatnot, but they aren't the same. You get stuck looking back and forth between the camera and the screen. The camera angles are weird and the receiving user is at the mercy of the person they're talking to as to what they're going to look at. Ironically, I've felt more personally engaged, more like I'm talking to a real person, having conversations in VR with robot-shaped avatars than I've ever had with human faces in Skype.