TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Oculus Research to Present Focal Surface Display Discovery at SIGGRAPH

212 pointsby srinathrajaramabout 8 years ago

18 comments

iblaineabout 8 years ago
VR sickness is primarily caused by latency. ie you move your head, the image takes a few milliseconds to respond and you feel dizzy. But there are other types of VR sickness, like the inability to focus on an object. This research improves your ability to focus on objects at different depths. Your vision is less blurry. So yes, this research does help eliminate nausea in VR. To say otherwise is misleading.
评论 #14359509 未加载
评论 #14359470 未加载
评论 #14363298 未加载
评论 #14359456 未加载
评论 #14359693 未加载
评论 #14359457 未加载
spyderabout 8 years ago
There is nice table in the paper which compares the capabilities of the different technologies trying to solve the DOF problem in HDMs.<p><a href="http:&#x2F;&#x2F;i.imgur.com&#x2F;8rdoeS3.png" rel="nofollow">http:&#x2F;&#x2F;i.imgur.com&#x2F;8rdoeS3.png</a>
mhalleabout 8 years ago
Having a display that can support ocular accommodation (selective focus by the eye) is an important research development, though it will most likely not change the viewer&#x27;s experience in a radical way.<p>Practical electronic 3D displays requires bandwidth reduction, both data bandwidth for transmission and optical bandwidth to create practical or lower-cost optical modulators. The goal is to use bandwidth reduction techniques that produce little or no visual artifacts. Some of the techniques used are the same as in 2D (spatial discretization, time multiplexing, compression), while others are unique to 3D (view discretization, limits on view angle, elimination of coherence).<p>Head-mounted displays are basically descendants of stereoscopes, the first 3D displays developed by Wheatstone in 1838. Wheatstone&#x27;s amazing discovery was that you can throw a huge amount of information about the world away, provide just two images from two viewpoints, project them out to infinity in front of a viewer&#x27;s two eyes using two lenses&#x2F;light paths, and a vivid sense of 3D is evoked. That&#x27;s an incredible amount of information reduction from real life.<p>In the traditional stereoscope, accommodation is thrown away, mostly because its really hard to recreate electro-mechanically, but also because we&#x27;re generally fine with it. Accommodation isn&#x27;t effective for distant objects (or for even larger depth ranges as we get older we lose our ability to accommodate), so we likely have neural circuitry to discount imperfect accommodation cues. One of the reasons we turn on bright lights when doing detailed work is to stop our eyes down and increase our depth of field, reducing the need for accommodation.<p>However, there have been perennial debates about the physiological impact of conflicting depth cues involving accommodation, and those debates are more interesting in VR where objects can be (virtually) very close to the viewer and the viewer can dynamically change their physical relationship with virtual objects.<p>Until you have a light modulator that can let you experiment with selectively modulating accommodation within a scene, you can&#x27;t provide real data on how important accommodation (even approximate accommodation) is for a particular application. Can&#x27;t wait to see the studies.<p>We did some similar focal plane manipulation in holographic video more than a decade ago, for related reasons (see Fig 7):<p><a href="https:&#x2F;&#x2F;www.researchgate.net&#x2F;publication&#x2F;255603167_Reconfigurable_image_projection_holograms" rel="nofollow">https:&#x2F;&#x2F;www.researchgate.net&#x2F;publication&#x2F;255603167_Reconfigu...</a>
评论 #14364433 未加载
DaveSapienabout 8 years ago
This is a much better (IMO)approach by Nvidia: <a href="http:&#x2F;&#x2F;www.fudzilla.com&#x2F;news&#x2F;graphics&#x2F;39762-nvidia-shows-off-its-light-field-vr-headset-at-vrla-2016" rel="nofollow">http:&#x2F;&#x2F;www.fudzilla.com&#x2F;news&#x2F;graphics&#x2F;39762-nvidia-shows-off...</a>
评论 #14359814 未加载
russdillabout 8 years ago
Would it be possible to detect the focal distance of the eye and change the entire focal depth of the display to keep it always in focus, similar to automated vision testing devices? It could then perform blurring of out of focus objects as a rendering step.
评论 #14360187 未加载
评论 #14360133 未加载
IshKebababout 8 years ago
I&#x27;m a bit skeptical of how much of a problem this really is. I have never noticed it in VR. Perhaps because:<p>1. Display resolution is still quite low, so really everything is blurry.<p>2. You will never be able to notice blurriness where you aren&#x27;t focused anyway because you aren&#x27;t looking there! Everything is always blurry in your peripheral vision.<p>3. Surely eye focus is a feedback system, like in cameras? I mean nobody has problems focusing on TVs because your eyes just magically change focal length until the image is sharp.<p>I am stereoblind so maybe it is a big problem for others.
评论 #14359918 未加载
评论 #14360191 未加载
评论 #14362270 未加载
评论 #14360313 未加载
Gravitylossabout 8 years ago
Why are holograms &#x2F; light field displays not technically possible now? I would assume think we have bright and dense enough displays, and can shape the microlenses.
评论 #14359600 未加载
评论 #14360666 未加载
评论 #14359376 未加载
评论 #14359519 未加载
6stringmercabout 8 years ago
Ctrl+F &quot;inner ear&quot; - no results. To me, that&#x27;s going to be the keystone to a functional VR experience. Until there&#x27;s some relatively non-intrusive mechanism to fool the body&#x27;s systems into playing along, I&#x27;m sorry, I don&#x27;t think image resolution or refresh or FPS will solve the problem. They&#x27;re all very important, sure, but I think the biology of the conundrum is the most challenging short term.
评论 #14359916 未加载
评论 #14359745 未加载
评论 #14360582 未加载
评论 #14362215 未加载
ryandammabout 8 years ago
I think this is overstated (though without looking through the prototype, this is only speculation).<p>During normal, outside-of-headset vision, we focus naturally and quickly on whatever we&#x27;re looking at. We don&#x27;t spend time with our eyes consciously defocused on subject matter in our foveal view. So anything that&#x27;s out of focus will tend to be in our peripheral view.<p>So this is a peripheral technology. I think everyone&#x27;s still looking for the killer additional tech that will make VR perfect -- but it&#x27;s not about one magic tech bullet, it&#x27;s about ecosystems slowly growing, and content getting better. (The headsets are better than people think.)
miguelollerabout 8 years ago
&quot;It may even let people who wear corrective lenses comfortably use VR without their glasses.&quot;<p>If just for this, it&#x27;s a move in the right direction.
评论 #14364194 未加载
eutropiaabout 8 years ago
Wont this also need gaze-tracking to be successful? In their video they described a manually moved camera.<p>Is this technology compatible with foveal rendering?
评论 #14361359 未加载
highdabout 8 years ago
Sort of strange to describe it as a &quot;discovery&quot; - I&#x27;m sure a team of engineers with a variety of fields of expertise spent 1-3 years solving problems that led up to this. A &quot;discovery&quot; would seem to describe something that existed in the aether prior to their work - it seems to diminish the innovation and effort they put in.
nsxwolfabout 8 years ago
Weird that I haven&#x27;t noticed this yet when using VR. I wonder if I&#x27;ll notice next time after watching this.
cobbzillaabout 8 years ago
i guess whatever the faults of the paper, i like that they do have a product on the market and are demoing and publishing research. magic leap? not so much.
AndrewKemendoabout 8 years ago
Honestly I found the paper critically lacking where they attempted to make reference or comparisons to virtual retinal displays. Saying that a VRD is functionally restricted to <i>moderate</i> FOV in comparison to the 120 degree FOV of the Rift - using only the embodiment of the deform-able membrane mirror as reference is ridiculous on it&#x27;s face.<p>Even a rough version of the deformable mirror AR VRD described by researchers at UNC Chapel Hill [1] accomplishes 100 degrees FOV with accommodation.<p>They went further with the Pinlight achieving 110 in 2014 [2]<p>The technical limit according to our own work for VRD FOV is H: 200o, V: 140o (Combined). So either they&#x27;re ignoring work in the field intentionally cause they don&#x27;t want to do VRD or they don&#x27;t know about it. My guess would be the former.<p>[1]<a href="http:&#x2F;&#x2F;telepresence.web.unc.edu&#x2F;research&#x2F;dynamic-focus-augmented-reality-display&#x2F;" rel="nofollow">http:&#x2F;&#x2F;telepresence.web.unc.edu&#x2F;research&#x2F;dynamic-focus-augme...</a><p>[2]<a href="http:&#x2F;&#x2F;www.cs.unc.edu&#x2F;%7Emaimone&#x2F;media&#x2F;pinlights_siggraph_2014.pdf" rel="nofollow">http:&#x2F;&#x2F;www.cs.unc.edu&#x2F;%7Emaimone&#x2F;media&#x2F;pinlights_siggraph_20...</a><p>edit: I find this whole thing extremely frustrating. Facebook could throw 2 billion dollars at VRD tech and actually get to a working stable consumer grade system if they wanted to - everything is there for it. Why aren&#x27;t they?
评论 #14360987 未加载
评论 #14362100 未加载
评论 #14360974 未加载
评论 #14360964 未加载
mattnewportabout 8 years ago
Article headline is misleading. Eliminating nausea is not the primary benefit of this research and isn&#x27;t even mentioned in the article. It may help a little with nausea in some cases but it won&#x27;t eliminate it.
评论 #14359816 未加载
评论 #14365116 未加载
sitkackabout 8 years ago
No one has mentioned magic leap.
jimrandomhabout 8 years ago
Nausea in VR is already a solved problem, at least for the case where you have a stationary camera. This is for giving people an extra method of depth perception, but one which isn&#x27;t strictly necessary.
评论 #14359542 未加载
评论 #14365124 未加载