Space Invaders in AR

Arcade Reality is a game, originally developed a couple of years ago for the Palm Treo, by Szymon Ulatowski, the man behind ToySpring, which has an interesting premise. For years, space invaders were slaughtered in video games by arcade dwellers around the world. Now, they stage their revenge in real world and the player must shoot them down via his/her mobile phone:

Recently Ulatowski has ported the game to the iPhone. While on the Treo it used some basic video registration to place the invaders around you, on the iPhone it takes advantage of the compass and accelerometer. Although it does make the alien spaceships a little wobbly, it still looks like the best game of its type currently available for the iPhone and a lot of fun:

It will cost you $3 on the appstore. More details can be found here.


Chri said...

hmm, i don`t know why people call this augmented reality ?!?!?
it`s just a 3D shooter game where the controller are the sensors of a device, there is no really need of realtime video import :(

it would be something other if the game is placing the enemies at real world objects, insteat of just the air.

mfg Chri

fbeeper said...

I always have loved this careful definition of Ronald T. Azuma ( page 2):

... AR as systems that have the following three characteristics:
1) Combines real and virtual
2) Interactive in real time
3) Registered in 3-D

It's 3D registration in real time! Otherwise, you've got to think that a simple projector pointed to a projection screen with a keynote can be augmented reality, isn't it? We are "augmenting the screen reality" :P

And, let me add another example... Anybody named PlayStation's EyeToy (1999) as augmented reality? NO! And it isn't augmented virtuality neither... you'll need full 3D registration on both cases, and EyeToy works on a 2D basis.

rouli said...

I agree that Arcade Reality, especially on the iPhone, is not "true" AR, since it's not registered to anything. However, I do feel a need to break free from Azuma's definition.
As far as this blog is concerned, AR is any technology that adds a layer of information onto the real world. Therefore, application such as Layar and Wikitude, that don't register and are not interactive, yet certainly augment reality, are AR.

Chri said...

The 1st point of this definition is simply not given in this game.

if you can play a game like this acrade rally in a complete dark room i can`t see any link to the real world ? neither information based (layar,,i`m from Austria its pretty useful for me :D ... ),nor enviroment based ( lot`s of AR mini, AR Transformers,...)

mfg Chri

rouli said...

On the Treo it's environment based (even more than AR transformers, since it registers to a real environment, not a marker). On the iPhone, it's less so, using a compass instead of image registration, and yet, I still think it's post worthy.

fbeeper said...

@rouli I definitely think this is post worthy, you make a very good work here :D Please, don't feel uncomfortable with my comment. :(

I'm very interested on the "new" view of the AR definition, and, I would like to discuss with you or the other blog readers if you don't care :D

Like I've pointed before, I'm a "lover" of the old Azuma's AR definition... but, YES, I'm also thinking that this definition isn't adapted to newer approaches and I'm interested on theorize about this definition :D

Your definition of "a technology that adds a layer of information onto the real world" is an interesting generalization of the AR term, but don't you think that it is too much wide?

To argue this I've got to retake my example of the "classic projector"... it is adding info to our reality (the projection screen) then... on a more generalized definition it might be AR, isn't it? (I'm not kidding, it could be. why not? it isn't a very impressive example of AR... but it could be AR depending on the definition)

But, I don't like very much the wider definitions... then:

I think that registration would be a very important key to name something as AR, don't you? Layar, Wikitude, ..., even Arcade Reality is making a registration: not a fully spatial registration, but (in this examples) a geolocalized one.
Then, I can imagine that AR needs registration. Not necessarily 3D, it could be only on some of all DOF. But I'm not sure if I'm considering all the possibilities. What do you think?

On the term of interactiveness in real-time I'm not sure that it is necessary... look, if I use AR techniques to augment one video on a preprocessing... for me it is AR, but applied on a different scenario. Do you agree with this?

And the concept of "virtual object"... is too narrow! I have liked the generalization of your definition, simply naming it as "information"

hmmm... that's all for now! I'll appreciate your opinions :D

P.S. Obviously I'm not a native English speaker... I've got to apologize for my awful English skills :P

Chri said...

hmm, i hope i understood you right with the "registering thing" ^^
For me AR the definition of AR there is no need that the applications is really registering a objekt, Augmented for me does only mean that you combine virtual information with the realworld , so like the app is already registering the real world in the way the application is looking where it is (gps/compas) so it`s putting the information to the right place.

I think there is a need of more detailed
partitionings of the technology differences.

im thinking of something like :

indirect controlled Virtual reality -> every PC with a TFT,CRT,Beamer, Overhead (just information on a screen where you can see with no userinteraction)

direct controlled VIRTUAL reality -> Arcarde Rally
(like IC-VR , but with userineraction)
indirect controlled AUGMENTED Reality -> Wikitude,,1st gen of marker based applications
(information placed over realworld, the user can only passive view the information or chance it via text editors ect.

Direct controlled Augmented Reality -> HandyAR,ACME, ARi (you know the japanese maid *g* ),2nd gen of marker based applicatoin ...
(like icAR, everything where the user can change attributes of a overlayed virtual object

for this special post its already ok that it is postet here, because its like a technology demo.

damn my english is also not the best, i hope you understand what i want to say :D

mfg Chri

fbeeper said...

Ok, 5 days later, I've got a second to answer this :D

And I've got one question, don't you think that Virtual Reality requires immersion? with an HMD or a Cave... ?

Chri said...

Yeah , i think this would lead to the general question what is Virtual Reality.
for me VR just a the definition of virtual "nontouchable" computergenerated content.

altough its quite old i think the following link explain it a little bit better (german)

mfg Chri

fbeeper said...

Ok, I'll take a look at it later, after taking a look to the demos & posters of ISMAR (and after a google translation :P)

But, even before of taking a look at it, is very clear that all terms like Augmented Reality, Virtual Reality, Augmented Virtuality, Mixed Reality, an so on... don't have globally accepted definitions. And I think that would be nice (or professional) to establish conventions and definitions of this terms, isn't it?

How can we do a good definition that would be acceptable for everyone? The most generalist definitions aren't the best for everyone, but the most specific definitions are too much closed and they haven't the many terms needed to englobe every approach...

I'm VERY boring, I know, but I think that this is like any other science and we need this exact terminology. If you have some ideas I'll be interested on them :P

rouli said...

I don't know, once you accept there's a continuum between reality and virtuality, you give up on having precise definitions.
The same goes for example with consciousness and unconsciousness. We would like to have dichotomy between the two, but the truth is the limit between them is blurred.

Post a Comment