The software would presumably know what you look like and can record your facial expressions and map it onto your avatar in real-time. There's technology now to create faces that never existed on a real person, and there's been technology for 15 years or so to do motion-capture of facial expressions and map it onto a computer-generated texture.

Plus I would hope that the big black glasses eventually give way to something sleeker, more akin to an eyepatch or sleeping mask.

For example,

HIGH-FIDELITY FACIAL AND SPEECH ANIMATION FOR VR HMDS Kyle Olszewski, Joseph J. Lim, Shunsuke Saito, Hao Li Siggraph Asia 2016 http://www.hao-li.com/publications/papers/siggraphAsia2016HF...

FACIAL PERFORMANCE SENSING HEAD-MOUNTED DISPLAY Hao Li, Laura Trutoiu, Kyle Olszewski, Lingyu Wei, Tristan Trutna, Pei-Lun Hsieh, Aaron Nicholls, Chongyang Ma Siggraph 2015 http://www.hao-li.com/publications/papers/siggraph2015FPSHMD...

Absolutely agree — our first biometric platform aimed to solve the “presence with half my face blocked” problem through crafty emg/firmware.

https://youtu.be/VUbJl_xiDFU

Also VR won't magically break through the speed of light, getting realistic telepresence with lag is going to be… interesting.

That is going to be the least of the issues. Even with a 100ms lag, conversation s still feel real time.

Try out high fidelity VR... they've nailed it already: https://highfidelity.com

Maybe a small spherical dome around your head?

(Could look like Google's nap pods but a little bit smaller.)