Slashdot Mirror

The folks at Mitsubishi Research actually came up with glasses that work like the ones in "They Live"... without the special glasses, you see one image, with the special glasses, you see another (secret) image.

Their paper is at
http://www.merl.com/publications/TR2002-011/

and the video is pretty darn amazing.

Power LEDs should be dimmable. Auto-dimmable would be ideal.

Here you go. That paper describes how to use an LED as both a light sensor and an LED. You could hack that together with a microcontroller and your existing case LEDs in an afternoon.

Elegant and powerful? Sure. But Readable? No way.

I like S-Expressions as XML replacement a lot, since for representing simple structured data its quite nice. But it just doesn't lead to very readable code when it comes to programming, [...]

Readable. Clean. Two of the most useless words ever used in a discussion about programming. They both reek of the kind of plain spoken "I know it when I see it" sophistry which tries to mask ones own petty preferences with a false sense of "natural" consensus. Unfortunately, this phrase, "I know it when I see it," has stuck in vernacular while almost no one remembers that Potter Stewart later recanted his view in Miller v. California, in which he accepted that his prior view was simply untenable. No programming language has ever been readable to anyone not already schooled in programming. No mathematical notation has ever been readable to anyone not already schooled in mathematics. This feckless obsession with syntax has got to be one of the biggest wastes of brain power since theologians debated how many angels could fit on the head of a pin. One of the examples of using the Lisp pretty printer published at around the same time as its introduction, if I my history is correct, was using the pretty printer to print a subset of Lisp as Pascal. Dylan already tried adding a more Algol-like syntax to Lisp. Paul Graham's Arc is yet another go at tacking some kind of level of extra syntax to Lisp. Qi has already added a sophisticated static typing system to Lisp. There is absolutely nothing new or interesting about any of this.

It's been very interesting to read this discussion. Thank for being so interested! A couple points:

1) I think Mason did a great job on the article. That's evidenced by the fact a lot of the posts here center on the questions I think are most interesting:

    a) Is there any way to balance the needs of society with the needs of the individual?

    b) How much information can you get out of networks of simple sensors?

    c) Are dense networks of simple sensors "better" than dense networks of cameras? sparse networks of cameras? networks of RFID readers? combinations of all those ...for various definitions of "better"?

2) That said, Mason's job was to write a very short article about a complex technology he only learned about last week, so it's understandable if some of the subtleties didn't make it through. If you're interested in the nuances, I encourage you to follow through the links and learn more (as many of are already obviously doing).

3) Discussions of privacy are very tricky in the absence of context. Privacy means something very different when you're talking about a citizen and their government, an employee and their boss, a parent and their child, or a lover and their spouse. If I am talking about one context and you have another in mind, then we're not likely to have a useful conversation. Please click through to this TR for a more detailed discussion: TR2006-005

4) It's worth pointing out that security was a good way to fund research these past few years. However the heart of the project is really about intelligent buildings, and how to make our lives better. It started with a focus on elevator efficiency and more recently has shifted to social networking and may yet move on to energy efficiency. Security was just one pit stop (but one the produced undeniably influential demos!)

5) Finally, the project is obviously not just two people, but dozens. Listing them all would have consumed Mason's entire column... so that's how it often is in the news. Many thanks to everyone who has been involved in the project over the years: AmbInt People

cheers,
Chris

It's been very interesting to read this discussion. Thank for being so interested! A couple points:

1) I think Mason did a great job on the article. That's evidenced by the fact a lot of the posts here center on the questions I think are most interesting:

    a) Is there any way to balance the needs of society with the needs of the individual?

    b) How much information can you get out of networks of simple sensors?

    c) Are dense networks of simple sensors "better" than dense networks of cameras? sparse networks of cameras? networks of RFID readers? combinations of all those ...for various definitions of "better"?

2) That said, Mason's job was to write a very short article about a complex technology he only learned about last week, so it's understandable if some of the subtleties didn't make it through. If you're interested in the nuances, I encourage you to follow through the links and learn more (as many of are already obviously doing).

3) Discussions of privacy are very tricky in the absence of context. Privacy means something very different when you're talking about a citizen and their government, an employee and their boss, a parent and their child, or a lover and their spouse. If I am talking about one context and you have another in mind, then we're not likely to have a useful conversation. Please click through to this TR for a more detailed discussion: TR2006-005

4) It's worth pointing out that security was a good way to fund research these past few years. However the heart of the project is really about intelligent buildings, and how to make our lives better. It started with a focus on elevator efficiency and more recently has shifted to social networking and may yet move on to energy efficiency. Security was just one pit stop (but one the produced undeniably influential demos!)

5) Finally, the project is obviously not just two people, but dozens. Listing them all would have consumed Mason's entire column... so that's how it often is in the news. Many thanks to everyone who has been involved in the project over the years: AmbInt People

cheers,
Chris

Somewhat off-topic, but the table in question is MERL's DiamondTouch, not Surface or a derivative thereof. The DiamondTouch predates Surface by quite a while - I got to use one at SIGGRAPH 2006. It uses an overhead projector onto the interactive surfac and pads that you must touch with some part of your body to use, generally by sitting on it. The table itself emits signals that are recieved through your body by the pad. This is in direct contrast to Surface and similar technologies, which use infrared emitters and cameras to detect touches, meaning no pads and a number of inputs that is only limited by the tracking software.

That being said, this still quite a neat application, and makes much more sense as a live monitoring tool than a wall of CCTV monitors. If you really wanted to get fancy, the bank of monitors could be hooked up to this system, displaying the most relevant camera footage at any given time.

This runs Linux, though it is not technically a MikeRoweSoft Surface product.

The product above is Mitsubishi's DiamondTouch screen. The folks who make it have released a Linux-compatible SDK.

You can read about it in the tech report. Although this system was not modifiable in real-time like the one in the article (older hardware), the model of the smart "lego-like" bricks was automatically slurped into Quake 2 and much FPS fun was had.

• "As a two-sided touch screen, the LucidTouch is a direct extension of our two-sided touch table, published previously as Under the Table Interaction (reference below).
  ...[snip]...
  Publications:
  Wigdor, D.; Leigh, D.; Forlines, C.; Shipman, S.; Barnwell, J.; Balakrishnan, R.; Shen, C., "Under the Table Interaction", ACM Symposium on User Interface Software and Technology (UIST), ISBN: 1-59593-313-1, pp. 259-268, October 2006 (ACM Press, TR2006-076)"

So is Apple copying Microsoft? Of course not. A table is not an iPod/phone is not a see-through tablet. As another replier mentioned, there's an obviousness about this "behind the screen" interaction. However, Apple fanboys (I'm not calling you one) like to think everybody's copying Apple.

• "As a two-sided touch screen, the LucidTouch is a direct extension of our two-sided touch table, published previously as Under the Table Interaction (reference below).
  ...[snip]...
  Publications:
  Wigdor, D.; Leigh, D.; Forlines, C.; Shipman, S.; Barnwell, J.; Balakrishnan, R.; Shen, C., "Under the Table Interaction", ACM Symposium on User Interface Software and Technology (UIST), ISBN: 1-59593-313-1, pp. 259-268, October 2006 (ACM Press, TR2006-076)"

So is Apple copying Microsoft? Of course not. A table is not an iPod/phone is not a see-through tablet. As another replier mentioned, there's an obviousness about this "behind the screen" interaction. However, Apple fanboys (I'm not calling you one) like to think everybody's copying Apple.

I had no clue what MERL was. It's Mitsubishi Electric Research Laboratories. Google turned up their web page easily enough. I recommend that anyone else not familiar with them peruse the 'Projects' section on the site. It seems rather impressive, at least with the few minutes I gave it just now.

Meh, Mitsubishi Electric Research Laboratory (MERL) has had a real product (DiamondTouch) in the hands
of researchers for over 5 years now:

http://www.merl.com/projects/DiamondTouch/

And even published papers on it. When will Microsoft do something original?

ftp://ftp.merl.com/pub/DiamondTouch/videos/Diamond Touch Multi-User Boxes.wmv (diamondtouch.merl.com)

As might the people at Mitsubishi Electric Research Labs, whose work in this area dates back to the 1990's IIRC.

Yea this stuff has been around for years: http://www.touchtable.com/site/index.php http://www.ms.northropgrumman.com/touchtable/ http://www.military.com/soldiertech/0,14632,Soldie rtech_TouchTable,,00.html http://www.merl.com/projects/DiamondTouch/ The Mitsubishi one can recognize multiple users. I've used it and it's pretty cool. Touch tables are nothing new but it would be cool to see Microsoft start marketing this to consumers.

Mitsubishi has a similar coffee table thing they call diamondtouch . It can handle two hands per user and the version I saw allowed 4 users. Oh yeah, and it was first publicized in 2001!

which I think Apple has the patent on.

Apple does not have any patents on the iphone. They have applied for about 300, but none have been granted yet. Regarding the multitouch interface, if you search the internet, you'll find that research has been going on in this area since the 1980's. At best, Apple might be granted a patent on the specific technology they've used to support multitouch in their touchscreen, but there are several other ways to accomplish the same thing, some of which are already available.

Can be found here:

http://www.merl.com/projects/DiamondTouch/DTflier. pdf

I wanted to post the specs itself, but thanks Adobe for continuously messing up their product: the copied text is put in the clipboard buffer backwards.

It's from the same group that brought you this project.

I agree HDTV on the wire could be a serious problem. But, what I've seen from Comcast (my only experience so far) it appears they're introducing extra compression, and the HDTV of a friend gives a status showing a transfer rate of 6MBs. But, this article shows HDTV needing aroudn 20MBs for streaming. To move to a world of on-demand HDTV for the masses would seem to (as they're claiming) require not only some prioritization of the network, but I would think it would also require a more capable internet, i.e., bigger pipes almost everywhere.

In addition, at my friend's, we found that HDTV streams could grind the house network to a crawl, I don't know if it's related (since it really isn't part of the network, but it is coming in on the same coax). Considering everything I've seen and experienced (hiccups in the picture, sometimes outright halting) I don't think HDTV over the wire is ready for prime time yet.

However, if I were a provider, I would have to consider that all of a sudden even a small percentage of my customers could consume all of my bandwidth and would have to come up with some approach to keep the pipes working.

Here is a video of Liquid War running on a MERL DiamondTouch: ftp://ftp.merl.com/pub/bogue/DiamondTouch/3-Liquid Wars.mov

Here is a video of Liquid War running on a MERL DiamondTouch: ftp://ftp.merl.com/pub/bogue/DiamondTouch/3-Liquid Wars.mov

MERL's DiamondTouch is a multi-user system and can tell distinguish between the touches of several users. If you're touching in one place and I'm touching in another, not only does the table detect both touches, but it can tell which of us touched which spot. Current DiamondTouch prototypes use an X-Y grid and so do not detect touches and gestures in a fully pixelized manner. This is a limitation of the current hardware, not the fundamental technology. You can read the original DiamondTouch paper here.

Also note that MERL DiamondTouch predates Jeff Han's work by about four years.

DiamondTouch uses capacitive coupling, but has a very different sensor architecture. It is closer to a radio receiver with a synchronous demodulator. See the following MERL technical report for more information.

http://www.merl.com/projects/diamondspin/

http://www.merl.com/projects/DiamondTouch/

http://www.merl.com/projects/diamondspin/

http://www.merl.com/projects/DiamondTouch/

Nothing special.
I saw a video of the diamondtouch (http://www.merl.com/projects/DiamondTouch/) system in my HCI lecture today.
Diamondtouch has been around since 2001.

These are almost exactly the same as the in-glass system I read about in a Mitsubishi R&D pdf a while back. linky, google html conversion it uses the table for power and data transmission, and detects liquid level through electric resistance measurements. neat stuff

It'd be rather neat if you could use two cameras to create a 3D-model of a person, instead of a 2D image -- some folks at MIT already did some research on this, but it'd be pretty cool to see this implemented in a game.

Maybe the Xbox 360 or PS3 could create a realtime 3D model on the fly?

Durability: All it takes is one pissed off kid stabbing it with a pencil to kill it. What about scratches? Assuming (hopefully) there is a clear screen protector most schools will wait until one can barely see through it before replacing. The screen protector would most likely cost a few hundred dollars and would need to be replaced once a year. Also repairing a big screen monitor like this is difficult and would require two people to pull it off the wall, deliver it to wherever it will be repaired and reinstall. Atleast three hours per person.

Obsolesce: Every few years these things get better and cheaper. $20k today is $10 in three years with a better picture and more features. In five to eight years these monitors will either sit in a pile like PII computers today or hang on the wall dead.

TCO: Between the initial cost, screen protectors and a short lifespan compared to a standard whiteboard these things IMO are way too pricey for the average secondary school.

Why not go with a LCD/DLP projector and a Mitsubishi DiamondTouch input device? A DiamondTouch "tablet" handles multiple, simultaneous input (two people can write on it at the same time), is incredibly durable and requires much less maintenance than a backlit screen. One could last for 10+ years handling input while the projector is updated every few years. IMO the TCO would be much lower than a huge touchscreen. As for durability it can be easily washed and very cheaply recovered. Since the sensors are on the sides and not behind the writing area it's rather immune from the "pencil penetration" scenario. Also Mitsubishi has been really good about driver support for GNU/Linux (along with MS Windows and Mac of course).

Will a backlit screen is nicer, a top lit projector and the above tablet IMO is a more realistic solution.

DiamondTouch Hardware
DiamondTouch Applications

Durability: All it takes is one pissed off kid stabbing it with a pencil to kill it. What about scratches? Assuming (hopefully) there is a clear screen protector most schools will wait until one can barely see through it before replacing. The screen protector would most likely cost a few hundred dollars and would need to be replaced once a year. Also repairing a big screen monitor like this is difficult and would require two people to pull it off the wall, deliver it to wherever it will be repaired and reinstall. Atleast three hours per person.

Obsolesce: Every few years these things get better and cheaper. $20k today is $10 in three years with a better picture and more features. In five to eight years these monitors will either sit in a pile like PII computers today or hang on the wall dead.

TCO: Between the initial cost, screen protectors and a short lifespan compared to a standard whiteboard these things IMO are way too pricey for the average secondary school.

Why not go with a LCD/DLP projector and a Mitsubishi DiamondTouch input device? A DiamondTouch "tablet" handles multiple, simultaneous input (two people can write on it at the same time), is incredibly durable and requires much less maintenance than a backlit screen. One could last for 10+ years handling input while the projector is updated every few years. IMO the TCO would be much lower than a huge touchscreen. As for durability it can be easily washed and very cheaply recovered. Since the sensors are on the sides and not behind the writing area it's rather immune from the "pencil penetration" scenario. Also Mitsubishi has been really good about driver support for GNU/Linux (along with MS Windows and Mac of course).

Will a backlit screen is nicer, a top lit projector and the above tablet IMO is a more realistic solution.

DiamondTouch Hardware
DiamondTouch Applications

It would allow players to move computer generate pieces as well... There's a reasonable picture of it in action.

I used one back at uni, they work well but I suspect aren't cheap.

It would allow players to move computer generate pieces as well... There's a reasonable picture of it in action.

I used one back at uni, they work well but I suspect aren't cheap.

In fact, the shadowing can sometimes be an advantage since you can often see things projected on top of your hands, which would be blocked in the rear-projection case.

In fact, the shadowing can sometimes be an advantage since you can often see things projected on top of your hands, which would be blocked in the rear-projection case.

http://www.merl.com/people/raskar/geomproj.html

Abstract

Projectors are currently undergoing a transformation as they evolve from static output devices to portable, environment-aware, communicating systems. An enhanced projector can determine and respond to the geometry of the display surface, and can be used in an ad-hoc cluster to create a self-configuring display. Information display is such a prevailing part of everyday life that new and more flexible ways to present data are likely to have significant impact. This paper examines geometrical issues for enhanced projectors, relating to customized projection for different shapes of display surface, object augmentation, and co-operation between multiple units.

We introduce a new technique for adaptive projection on nonplanar surfaces using conformal texture mapping. We describe object augmentation with a hand-held projector, including interaction techniques. We describe the concept of a display created by an ad-hoc cluster of heterogeneous enhanced projectors, with a new global alignment scheme, and new parametric image transfer methods for quadric surfaces, to make a seamless projection. The work is illustrated by several prototypes and applications.

Is here

See conversational Penguin robot.

The technique is fast and the results are impressive.

This is a paper from Siggraph 2000 about interfacing computers with modeling clay and lego style blocks:

http://www.merl.com/papers/TR2000-13/

This is not a new invention. Mitsubishi Electric has done this before: iGlassware

The Japanese version does not require batteries in the glasses or pitchers.

The same problem can also be solved by measuring capacitance of the glass across the remaining fluid. (I don't really understand this, but I'm believe it's fairly simple.)

The article references this, in fact.

Are you sure you don't want that?

The first example of this I saw was Diamond Park in 1995, a 3d virtual world from Mitsubishi ER Labs that you navigated on a bicycle. Then Georgia Tech published a bike race virtual environment based on the olympics in '97, that project is still going. My gym at work has 2 Tectrix VR exercise bicycles with video monitors and left/right turn by leaning that give you the option of a winter tour, island tour, or basic games, you can do any of the above with 2 players, and you can set it to pay attention to hills or not... I find that using it instead of the regular exercise bike I will get a lot more violent, as if I were playing a real sport, a lot more cardio workout, and it's fun when you go to the gym with a friend, whereas with the other I just plot along reading my magazine or book... but they don't make them anymore, according to the gym, so when they break, we're SOL. www.bodyquest.com claims to still sell them, but I think their website is just out of date.

The basic notion is this: we regularly spend time physically near people we are related to. We can thus go the other way and use physical proximity data to infer relationships between people. For example, during weekday business hours, I am near business colleagues, while at night and on weekends I am near friends and family. By simply observing these patterns of who I am near, when, and for how long, the system can infer the types of relationships I have with the people I encounter (for example, it can distinguish between business and social relationships).

The first application of this idea was Social Net, a system to introduce users to new people. Social Net notices patterns of physical proximity over time (i.e., frequent and/or long encounters) to infer that two people are engaged in a similar activity (or that they share interests). If the two people don't know each other, Social Net looks for a mutual friend (ie, someone who knows both). If a mutual friend is found, he/she receives a message suggesting the two people should be introduced.

Using physical proximity data, over time, allows Social Net to infer shared interests between people without requiring them to identify what those interests are. It has the nice property of filtering out chance encounters with people on the street, since we it considers the duration and frequency of an encounter. Finally, the mutual friend brings accountability into the whole process, so that your device is not telling you to go up to a complete stranger and introduce yourself.

There are other neat things you can do with this data. For example, an app can infer business vs. close interpersonal relationships, then attenuate a cell phone's ringer when the user is near friends or family (since it will be able to infer that the other people are friends or family based on past histories of physical proximity). An app could also automatically exchange music lists between people as suggested by Korteum, but use patterns of physical proximity to infer shared interests, rather than requiring users to manually enter those interests in. The real hurdle to getting these apps out there, of course, is finding the one killer app that makes people less wary of transmitting presence information into the environment.

Uses shutter glasses, but instead of displaying alternate Left/Right images and alternating which lens is blocked for 3D, it opens and closes both lenses in sync and displays the image you want to see when they are open and the NEGATIVE of the image when they are closed. You just see your image a little dimmer, everyone else just sees a grey blur. Also can be used to display the difference between your image and another (when the lenses are closed) so that people without glasses see an entirely distinct image.

I "came up" with this idea about 3 months ago, probably because I had read about it previously.

I've been meaning to get this on Slashdot for some time now, but I worked on a much more powerful version of such glasses over a year ago at Mitsubishi Electric Research Labs (MERL) in Cambridge, MA, with Researcher William Y. Yerazunis. Here's the technical report if you'd like to see it. We also filed a few patents way back then as well, so I wonder if this work infringes on our own.

We can actually hide secret images within any image or animation you'd like, not just an obvious blank screen. We also designed a cryptographically secure version which isn't cracked by simply having another pair of special glasses (you also need the private key). Check out the paper, it has some image examples (there might be a few technical errors in it that we later fixed but wasn't updated in the paper. I'm not at MERL anymore, so I haven't bothered checking really).

Also, we made a video demo for the conference which our technical report was accepted in paper form (at OzCHI2001). I have that video, and can digitize it if there's enough demand. By the way, while I was testing the glasses, I actually used They Live screenshots so that one could simulate Rowdy Roddy Piper's shock upon seeing the billboards and aliens. Also, we referenced John Carpenter in our paper.

We've been able to sense a finger from several inches away with one of these that has been adjusted correctly. If a person can move a finger up to an inch, even without being able to apply pressure with it, a sensor like this will have no problem detecting that.

(Yes, this is a tech report about the Mitsubishi Electric "Smart Drinking Glass" that was reported earlier on slashdot.)

The coolest part is that you can download your model into Quake and frag your friends inside the structure you built.

There was a paper on this at Siggraph 2000 and a tech report is here.

The coolest part is that you can download your model into Quake and frag your friends inside the structure you built.

There was a paper on this at Siggraph 2000 and a tech report is here.