Homeless people and wrong expectations for AR glasses

Today just some mixed thoughts on AR and society, guess it’s kind of a blue thursday.

People start discussing augmented reality glasses and their impact on society. Once, everyone has a head-mounted-display (HMD) connected to broadband internet, the world will completely change! 2010 is already being hyped as changing the world. But hohoho, media, slow down please!

John Carpenter - They Live

Charlie Brooker from the guardian just ripped on augmented reality and the possibility to get rid of homeless people in your eye sight through augmented reality (in this article). It depends on your style of humor if you like his ironic comments. But people keep forgetting that we are still a long way from a perfect overlay! And why do we have to stress these silly negative scenarios? If he would have overlaid the homeless with suit-wearing equals, rather than cartoon characters, he might even start a conversation with a guy from the street “by accident”, thus emancipating/”equalizing” the homeless and bridging social gaps. Sounds better now, hu? ;) Naa, it’s still bad.

Regarding the realistic visual AR deception: there are not only broadband bottle necks or realistic rendering issues to manage today, but much more. The tracking needs to be perfect, the lighting needs to be correct in real-time – and really the biggest problem: the field of view of any HMD is pretty narrow and the hardware is insufficient. The problem comes with the two human eyes and the projection plane. We still don’t have the convincing solution at hand. If we are observing the world through glasses and two cameras (for each eye and thus stereo view) it wil still not be realistic for a long-time period. If the field of view is bigger, that’s only half the way. We experience the problem with all these stereoscopic movies again: the projection plane (the screen) is not the actual focus plane of our eyes, thus giving a headache, plus the HMD cameras don’t toe-in or converge, thus giving us a non-realistics impression of the world and leading to stereoscopic headaches and wrong images again that are out of sync between brain and eyes. Well, I could go on…

Anyway, enough technological mumbo-jumbo, the point is: realism with glasses is a development I’d love to see, but there will soon be no way of tricking people into believing the presented image by 100% through glasses. We can and will gain this effect much sooner with desktop/handheld devices, but not convincingly on our noses for the next 5-10 years, imho. Thus, all these scary cyberpunk scenarios popping up, are way to early. Of course, they sound good in an article, I have to admit.

The only current issue we have to be worried about is: people living among other humans in a society get lonely or by their own will disconnected with everybody around – in favour of connecting to other prefered (virtually present) people. But of course this just continues the series of walkmen, cellphones, mobile gaming devices and is not a problem per se, but needs to be considered by each individual in each situation.

Funny applications to socialize try to counter this development of disconnecting-with-the-real-world, which is actually kind of silly. But that’s how it has always been. One technology may drive us apart, another technology is needed to bring us together again. Here, I just want to appeal to the people’s judgement to focus on the right issues regarding technological development and don’t go buzz wording around. Technology might even be an ice breaker and of course may work to discover new people around you, that share one of your interests, etc. There is a whole bunch of new possibilites, if we do it right!

We are still in the early stages of AR, so let’s don’t do the second step before the first. So, don’t get me wrong, I’m eagerly waiting for Vuzix and other companies to launch their HMDs for every Joe Sixpack. I will be among the first to scream like a child on Christmas once I walk with it through my office. :-) The next thing is a floating screen and HUD for every situation. But we might focus on the implications for these scenarios rather than the ones by realism.

Talking about realism… until we reach a scary future with invading Aliens and reversed augmented reality like in John Carpenter’s They Live it’s still a while to go… But once it’s there, we might be unable to live without it anymore. Quoting from the movie: “wearin’ these glasses makes you high, but, oh, you come down hard“. :-)


  • Hey admin I was wondering if I could steal this for my site. I will provide full credits to you and your site. Thanks.

  • Pingback: ???? | ????? ???? » ????????

  • Hi Tobias,

    Do you know the state of the practice in regard to projecting, on the inside of glasses, imagery that is mathematically transformed into the shape that it would have at the distance of the glasses from your eye, were the imaged object actually existing, say, two meters away?

    That is… place a real object at two meters and look straight at it, wearing glasses. Now, raytrace from the object to the lens of the eye, and capture the traces at the surface of the glass in front of that eye. Remove the object, and project the captured image on the glass. Does that resolve this issue: “the projection plane (the screen) is not the actual focus plane of our eyes”?

    Danke, Dave

    • @david: I’ll get back to you by mail.

      @Brooker: more text and less questions marks would be helpful. I’m sure there will be glasses out before that date and I want and will use those. but it will only be a HUD: additional information, that are location-aware. but we will not have a full AR-impression, where we can’t tell the difference between CG and REAL.

  • @david: just send you my thoughts on the lens/eye/camera/screen-problem.

    if this of general interest maybe I start a discussion on it soon?

  • Toby,

    it’s certainly of interest to me :)

    It is clearly cheaper and easier to project on one lens than on two. I imagine avoiding the complexities of eye tracking and convergent stereoscopy.

    I wonder about “a discrepancy between where your eyes actually focus and where your brain thinks it has to focus.” I offered an example of projection for eye focus at two meters because of the tremendous depth-of-field of the eye… viewing objects further than reading and nearer than, say, 50 meters might very well be augmented by text projected as though at two meters. Further, the “toe-in” movement of both eyes to converge where “your brain thinks it needs to focus” might occur unchanged given monocular overlay, reducing or eliminating oculomotor disparity.

    I am a novice, however, and hoping for insight shared by experts.

    My background is in software engineering, visual media, and entrepreneurship and I am in the process of proposing a book on mobile AR on Android.

  • @Toby & @david: The issue addressed here has also come up in my research into AR (startingpoint = architecture & urbanism). I am really curious what will happen once we get good transparent screens and based on those, maybe even contact lenses for commercial use. I do agree that focal point issues and the complexity of the eye’s working is something we won’t be able to work out overnight. However creative workarounds might popup soon.

  • Substantially, the article is really the greatest on this deserving topic. I fit in with your conclusions and will thirstily look forward to your upcoming updates. Just saying thanks will not just be sufficient, for the wonderful lucidity in your writing. I will directly grab your rss feed to stay abreast of any updates. Good work and much success in your business dealings!

Leave a Reply