From: hlr@well.sf.ca.us (Howard Rheingold)
Subject: Re: Viewing 3D images in a Head Mounted Display
Date: 16 Jan 92 18:37:35 GMT



cs_d476@ux.kingston.ac.uk (pete leaback) writes:

>While using W Industries "Virtuality" head mounted display, I found
>it hard to fuse the two projected views of an object into one 3D object.

>I presume what was happening was that my eyes were refusing to go 
>cross-eyed enough to bring the object into true.

>(stuff deleted)

>A head mounted display is trying to get ones eyes to do something
>it is not used to, ie focus at one distance and go cross-eyed for
>a different distance. If the object is far from the viewer, this is not
>a problem, but close up (where we want a 3D effect!) it may be. 


Indeed, "convergence" is what is required for full 3D effect, and that
doesn't just happen. Bob Stone in the UK and Susumu Tachi in Japan are
both working on automatically-converging slave-cameras. From what I
have heard, the Virtuality system doesn't include a convergence
mechanism.

From my limited understanding, convergence means that your eyes indeed
become slightly but crucially "cross-eyed" when you pick an object to
focus on. The mechanisms in auto-focus cameras have been adapted to
convergence systems by one of the contractors working for Bob Stone.
And will somebody in the UK please kick some ass and get Stone's group
connected to Usenet?

Howard Rheingold