I could have used something similar, when doing XR-ish laptop human interfaces. Assuming sufficient resolution. Knowing screen pose relative to eyes, from hea

I could have used something similar, when doing XR-ish laptop human interfaces. Assuming sufficient resolution. Knowing screen pose relative to eyes, from head tracking on the laptop webcam, it would have been nice to simply derive keyboard pose, without needing to track it separately (with an extra camera/mirror/kludgery).