The Advanced Human Interface Group (AHIG), MIT Media Lab. The ICONC System, demonstrated by Hrafn Th. Thorisson, Summer 1993. The system enabled the use of co-occurring, natural speech and gesture to interactively describe the arrangement and movements of objects in a room. The computer would interpret the user’s actions and figure out which objects the user was talking about and how to arrange them based on spatial information in the user’s speech and gesture. [Excerpt from the YouTube description, continued below]

[If you can't see the video, click here to visit the post]

The main authors of this work were David Koons (spatial knowledge, multimodal integration) and Carlton Sparrell (gesture recognition), directed by Richard A. Bolt. This technology is described in part in the paper “Integrating simultaneous input from speech, gaze, and hand gestures” by D B Koons, C J Sparrell, K R Thorisson (1993).

UPDATE (Oct. 30th, 2009): The article stated, wrongly, that this took place in 1994. This has been corrected.

Links & references

• Kristinn R. Thórisson’s website at Reykjavik University & at the MIT Media Lab
• YouTube page for this video

Layar 3D makes use of OpenGL, the accelerometer, the GPS and the compass of the phone. Developers can place 3D objects in their content layers based on coordinates. Objects can be optimized in size and orientation to create an immersive and realistic experience. The 3D capabilities support live downloading and rendering of 3D objects. Actions such as “open link” or “play music” can be assigned to 3D objects. [Press release]

Looking forward to early results from the minds of their developers.

Embedded videos after the jump

Boeing 747 flyby

3D tagging in augmented reality

Arcade game animation (Pac-Man)

Links & references

• Layar’s main website
• Layar’s 3D demo page
• Layar’s press release (Sept. 22nd, 2009)

Correct prediction

Undetermined

Wrong prediction

Today we’re launching a special page to store past and present predictions regarding future technology developments. At the moment all are in the area of augmented reality. Below is a list of new predictions; the complete list can be found on the new Predictions page. The page can also be accessed through its link in Think Artificial’s header-menu.

What do you predict? =)

Year 2009

Initial iPhone development and app popularity

• Prediction on August 28th, 2009: In October 2009 many developers are reported working on augmented reality apps for iPhone. This is partially due to the success of 5 wildly popular apps in the iPhone App Store.

Year 2010

Rise of serious augmented reality apps

• Prediction on August 28th, 2009: In January, 2010 interest in fluff (useless, silly) applications is declining. Much more powerful, serious AR apps are on the rise.

Major stores adopt augmented reality

• Prediction on August 28th, 2009: In January, 2010 the first major store announces mobile AR support; possibly an app that indicates product locations in shelves, or one that shows information about products. There are rumors of at least 3 other stores preparing a launch.

EDIT: The following two first appeared with dates, but were changed within an hour to mid/early/late -month notation (I’m not psychic! =)

Open-source augmented reality platform

• Prediction on August 28th, 2009: Mid April, 2010 an important open source AR platform standard is introduced.

Augmented reality eyewear & head mounted displays

• Prediction on August 28th, 2009: Early May, 2010 the first specs specifically designed for mainstream, mobile AR are announced. Use of eyewear-screens (head-mounted displays, AR glasses) increasingly popular among young and young adults.

Before September 15th 2009 one of the following two events will occur:

1. Apple releases initial support to iPhone augmented reality applications.
2. Apple announces an extension to the the iPhone SDK, specifically intended for augmented reality app development.

[Posted on Jul 13, 2009]

Only eleven days after the prediction news began rushing in; among many others reporting, MacRumors said on July 24th 2009:

The L.A. Times reports that Apple will begin allowing developers access to the tools they need to produce augmented reality applications starting with upcoming iPhone OS 3.1. [So far, AR applications] have used unpublished APIs which prevent them from being allowed on the App Store. Apple, however, told one developer that the tools necessary would become available with iPhone 3.1. [MacRumors]

In short, Apple is releasing their initial support to augmented reality applications. The Los Angeles Times posted the article that broke news that Apple told developers of the Nearest Tube AR train finder (Acrossair) that augmented reality apps will be allowed in the iPhone App Store in September, as predicted… let’s see if it turns out to be September 15th

Stay tuned for more predictions.

[NOTE: Above is a snapshot for July '08 to July '09;
See current 12 month trend]

Officially, Apple’s iPhone SDK does not offer access to any APIs for manipulating live video, forcing developers to use the available but unsupported ones instead. That’s a shame because as you can see, there are a lot of unique concepts out there for implementing Augmented Reality on the iPhone. [...] However, there’s still hope. According to some hearsay out there, Apple is interested in enabling these types of apps. [ReadWriteWeb]

There’s no question that Apple realizes the potential of AR and their position to dominate the market. The iPhone’s media attention and hardware power can make for a Walt Disney of augmented reality.

Having correctly predicted the type- and date of debut of the first mainstream AR application, the oracle speaks again—let’s see if I get this one right:

EDIT (August 13th 2009): This prediction was correct. Only 11 days after I posted this prediction it was announced that Apple’s iPhone OS 3.1 will support augmented reality applications—expected release is in September (as predicted). See further details in Correctly predicted: Augmented reality support in iPhone OS 3.1

Will augmented reality on the iPhone accelerate development of wearable displays?

The iPhone did the impossible and brought back head-mounted displays (HMDs)—often known as virtual reality goggles. Many modern HMDs are now relatively cheap—the pricetag is no longer long enough to use as a belt—and, on occasion, the design is sleek. (They still suck, but keep improving.)

Using phones’ screens for augmented reality applications is literally like peeking through a keyhole. Demand for alternate interfaces increases as a result of mainstream AR, and the greatest of these are wearable displays; AR specs (see-through displays, shaped like glasses). These have been around for decades (image shows MIRAGE - undergrad student project from 2002) but are still expensive as hell. Will the iPhone do for AR specs what it did for HMDs?

Importance of AR+wearable display interplay

The combination of wearable displays and augmented reality has countless uses. It’s important, in fact, with potential to revolutionize machine interfaces and manufacturing practices. Picture the following, for example: Desktop computer output (screen) can be overlaid onto reality like any other AR graphic, displayed in 3D as if they are sitting on a desktop.

Instead of buying two screens for a dual-screen setup at home, a student on a tight budget could buy AR specs. Through these he can display as many virtual computer screens as needed. If quality were to improve and price decrease, AR specs could serve as flexible, space saving, easy to transport and cheap alternative to any modern hardware display.

Augmented reality is not a type- or collection of applications. It’s a type of medium. In today’s economy, requiring cost reductions and preservation of resources, a future where a single gadget can replace several seems ever more likely. The boom of augmented reality is more probable than ever.

Visual comparison of Layar & Wikitude

Comparing the first mainstream AR application to the second, Wikitude to Layar, there’s a noticeable improvement in user interface. The radar looks cool, but is it just a compass?

And FINALLY someone implemented layers. It’s what it sounds like; visual data is organized on layers that can be made invisible/visible. Like Photoshop. And the infinite number of layer’s are for sale to companies with good ideas (—that’s their business model).

Layar’s video shows examples for a real estate guide service (good choice for an early layer—encapsulates the idea, potential). They’re currently requesting layer ideas from companies, but I’m sure they’re working on making people able to create and share their own layers. That’s what I would do.

The videos embedded below show Layar in June ‘09 and Wikitude in October ‘08 (pre-release).

Layar promotional video (released June ‘09)

Wikitude demo video (released October ‘08)

Layar’s not the world’s first browser—it’s the first augmented reality web

The world’s first augmented reality browser for mobiles. So goes Layar’s tagline. But, uUnless I’m mistaken, they don’t actually mean it’s the first augmented reality browser for the Web (as in the World Wide Web). It’s the first browser for the web they developed, an augmented reality web! So the browser is a first, but only because the web it browses is a first. So the tagline isn’t technically wrong, it’s a marketing mistake perhaps, but not wrong. The Augmented Reality Web aspect is the cool part and should be scream of the brand. Including the term “browser” in there just confuses things. But nevertheless, they’re getting full media coverage.

Layar is high on my list of “AR companies that get it”—understand one of the unique opportunities in mainstream augmented reality. Respect where it’s due. They’re developing a well designed product with potential, deserving of success.

Hat tip to the SPRXMobile team; Raimo van der Klein, Claire Boonstra, Maarten Lens-FitzGerald. Well done.

Links & refs

• What people are saying right now about Layar (Twitter)
• Official Web presence: Website, Twitter user, Android Market entry

The US Postal Service has begun using a pretty neat online augmented reality (AR) tool. Developed by AKQA/DC, the Virtual Box Simulator is a marker-based AR applet that enables you to use a webcam to find a box size and shape that fits a shipment item.

Print out a marker on paper and point your webcam at it: a 3D virtual box appears. Their video example (click ‘See how it works‘ on the right-hand menu) shows the size of the box and transparency being altered so that when you place your shipment on top of the paper its easy to see which one best fits.

Compared to similar applications, even though it only shaves a minute or two off your time at the post office counter, its still the most useful I’ve seen.

[via Adrants]

[If you see no video then click here to visit the post]

Earlier this month reports of a “robot scientist” made their rounds telling of a robotized lab and AI system that generated its own hypothesis, ran experiments to test them and with the process “discovered new functions for a number of genes in Saccharomyces cerevisiae, aka brewer’s yeast.” The robot, called Adam (cute), is being developed by Aberystwyth University and the University of Cambridge in the UK.

Then again this week there are news of what’s essentially a genetic algorithm that extrapolated the law of conservation of momentum, and Newton’s second law of motion from a pendulum’s swing; and without knowledge about physics and geometry.

A good week for the machines! Happy holidays =)

Created by int13 the game is marker based (meaning that the software uses pattern on a paper to know how the world is oriented) and features some cute cuddly Pokémon-styled creatures that you train and control (to fight). See the embedded videos below.

Augmenting reality to ignore reality

Love of the AR concept leaves me blind to shortcomings, but my girlfriend stayed vigilant and posed an observation: How does AR enhance this particular game? And she’s spot on.

Other than providing a flat surface, the real, physical environment plays no part at all in Kweekies’ gameplay. Making that surface virtual would provide more or less the same experience and remove the need to lug around a fragile piece of paper with inkblots on it (can it even be folded without producing errors?). But it does look cool, I’ll give it that. The portal concept and design is nice even though it looks like it’s just bling. Maybe Valve could do something functional there.

Is augmented reality in virtual reality’s footsteps?

This does beg the question whether primitive AR implementations will cause people to write them off as ‘cool… but pointless‘, thereby slowing down AR development. After all, that happened to virtual reality in the 1990’s: creating viable consumer products was impossible but we still made some primitive demos that spread hype like a nuclear shockwave. Time passed. The only things that remained exciting were machines with sky-high pricetags that even game arcades could hardly afford. Public demand decreased and VR R&D crawled out of sight and into the shadowy corners of labs, hospitals, military bases. By the time we had everything to make killer consumer products (ca. 2005)—people had lost all interest.

But maybe the masses will love all AR applications regardless? They’re cool, there will be others like Wikitude that’ll do practical stuff. One of the most popular iPhone games is a ‘yo momma’ joke generator. So who knows?

[Kweekies via VR Geek Blog]

The notification came yesterday and what a delightful wake-up call. I knew there’d been about a thousand submissions last year, and that they’d had to make a delay this one due to an increase, so it was quite unexpected. (Each category is drastically different—but it still makes my head spin to be on a list with Ars Technica! Awesome.)

My thanks go out to the judges, bows to the fellow finalists and best wishes all around!

Only yesterday the police employed gas to disperse a crowd for the first time since 1949 when Iceland joined NATO. The people are protesting the government’s poor management in the months since the collapse.

I’ve been brooding on ideas of how to introduce innovation as a key component in the new society. This is not an easy thing to do—Iceland is drenched in history-based economic models to make decisions. When that model goes too far, as it has royally, it does not favor innovation, startup companies and extensive research; businesses tend to put their money where its safe, not into the abyss of new ideas that may- or may not revolutionize the world.

This compels me to do something; try to help steer this country into something less extreme, more innovative. I’ve been doing presentations about how special, unique and rebellious ideas can bring about great things if done right. There’s much more planned for 2009, and also: I’m learning about politics—can you believe that?

There’s a lot on my mind—but Think Artificial is not dead. Or if it is, then it’s only because there’s the new Think Artificial of 2009 waiting to take its place. I hope to see you and your comments around. I humbly thank you for reading in 2008 and for sticking around!

Best regards,
-Hrafn

Read on for videos and specs.

Video and specs

The video (see below) shows the precision and awesome dexterity of Motoman as he puts together a dismantled camera, much to the dismay of a tired and confused Japanese actor put there to emphasize how awesome the robot really is. Chefs and musicians beware: Motoman can also play drums and cook.

The robot is being developed by Yaskawa Electric Corporation and is pending commercialization. They began taking orders on December 1st 2008 and expect to sell 1,200 units in 2008, 2,400 units/year for 2009, 3,000 units/year for 2010 and after.

Motoman is 135 centimeters (4.5 ft) tall and weighs 220 kilograms (480 lb). It has 15 joints: 7 in each arm (similar to humans, I believe) and 1 at the center of its torso.

Links & references

• Yaskawa Press Release
• Motoman Humanoid Robot Cooks Okonomiyaki With Manual Precision (Bornrich.org)