From poup@mic.atr.co.jp Sun May 10 11:29:39 1998 Received: from burdell.cc.gatech.edu (root@burdell.cc.gatech.edu [130.207.3.207]) by lennon.cc.gatech.edu (8.8.4/8.6.9) with ESMTP id LAA09819 for ; Sun, 10 May 1998 11:29:38 -0400 (EDT) Received: from wheaten.hitl.washington.edu (8eGD7sinZ/Z09JduK2waJ5UH1j5iItRF@[128.95.73.60]) by burdell.cc.gatech.edu (8.8.4/8.6.9) with ESMTP id LAA15154 for ; Sun, 10 May 1998 11:29:36 -0400 (EDT) Received: from mailhost.mic.atr.co.jp (mic.atr.co.jp [133.186.20.201]) by wheaten.hitl.washington.edu (8.8.8/8.6.12) with ESMTP id IAA28404 for <3d-ui@hitl.washington.edu>; Sun, 10 May 1998 08:29:30 -0700 (PDT) Received: from pop.mic.atr.co.jp by mailhost.mic.atr.co.jp (8.8.8+2.7Wbeta7/3.6W) id AAA08457; Mon, 11 May 1998 00:28:57 +0900 (JST) Received: from mic.atr.co.jp by pop.mic.atr.co.jp (8.8.8+2.7Wbeta7/3.6W04/07/98) id AAA01890; Mon, 11 May 1998 00:28:57 +0900 (JST) Message-ID: <3556A788.C2665B72@mic.atr.co.jp> Date: Mon, 11 May 1998 00:23:52 -0700 From: Ivan Poupyrev Organization: MIC Labs, ATR International X-Mailer: Mozilla 4.04 [en] (Win95; I) MIME-Version: 1.0 To: 3D UI list <3d-ui@hitl.washington.edu> Subject: Re: HMDs, CAVEs, COVEs, monitors, oh my! Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Status: RO > >Matt Conway and Jeff Pierce wrote: > >In terms of cost and sheer numbers, I believe that most 3D UIs will > >appear > >on mundane setups, those with a humble monitor plus mouse. These will > >outnumber caves, coves, reactive workbenches, head mounts, interactive > >holograms (cool MIT stuff!), or even Very Large Screens for the > >forseeable future. > > I'll buy that we probably won't see 6 DOF input devices on the desktop any > time soon, but it'd be nice to at least get a 2nd input device for our > other hand. I am wondering which interfaces and applications can be called 3D and which can not. When I am playing Doom on PC with a keyboard, is it 3D interface or not? If I develop some extremely cunning interaction techniques for flying through 3D space using only keyboard, would it be a 3D interaction techniques or not? If not, then what kind of techniques it would be? I believe that although desktop PCs are not inherently good for 3D UI, they are important from simple economics perspective: there are millions and millions of monitors installed on the desks all over the world. What should we do to make 3D user interfaces convenient enough so that people use them on simple PCs without adding specialized input devices? Is it possible at all? That would be an interesting search. In relation to 6DOF input devices for desktops, one of the interesting approaches is camera-based tracking. On the last IEEE Conference on Automatic Face and Gesture Recognition (this April) there were some impressive results. For example there was a guy from Australia, who was showing real-time gaze direction tracking using PC camera. He told me that he can get 1 mm accuracy in tracking, taking that the camera is close enough to the face, his video demo was really good. Mitsubishi was showing their artificial retina chip, an imagecapturing chip that can 1) extract edges 2) run correleations 3) do it bloody fast 4) cost $5 a piece. Using 2 cameras with this chip they were able to track head orientation. In real time and it was good. Well, at least on thier video tape. >> Doug Bowman and Jeff Pierce wrote: >> That's why I said that the mini-CAVE doesn't buy you much more than >> an expanded display area. Of course, you can do stereo, but you >> can do stereo on a monitor as well. > > I think there's actually an interesting question here: how large does a > monitor have to be before your motor perceptual system stops telling you > that you're looking at a box and starts telling you you're looking through > a window? Or is there something else we need to do besides size? I'm sure > head tracking is probably a part. Would embedding the monitor in a wall > help? Putting a picture frame around it? Thoughts? I have been playing a with a big back projection screen, which was about 8 feet by 8 feet in size (monoscopic). It seems that one of the important factors is a relative scale between the size of the output surface and your body. When you can stand in front of the 3D CG objects, it feels like they exist in the same space as you. The main problem of course is head tracking and that it is stack in one place: you should always look at it to do anything. Resoultion seems to be another important factor. Either Matsushita or Mitsubishi is producing virtual fish tank, which is a high definition TV shaped as a fish tank showing you fishes floating inside, not CG, just a looped video I think ... If you stand just 8 feets away, it is really difficult not to believe that fishes are real. I guess one of the reasons is that with the distance our sterescopy depth cues are getting worse, especially for small objects like fish, and we base our depth judgements on shading, occlusion and etc. where quality of image is crucial. So, if we want people to get really immersed into CAVE or projection screens or COVE (I do not mention HMDs) we have got to get much better resolution then current NTCS resolution that is often used in this sort of setups ...(by the way, does anybody knows the maximum resolution of the video projectors that are used in CAVE type environments?) It seems that plasma TVs are going to be next step in monitors systems. They are on sale here in Japan and expensive but, boy, they are big and thin and, unlike conventional monitors, it is possible to make them as big as you want! I think that 2010 scenario, which Jeff mentioned (that we will be getting cheaper, bigger and thinner TVs after digital broadcasting gets in place) seems to be much closer. Ivan