Idea for the future: Augmented Reality models

[RageOfAges]

OH MY GOD that is one of the neatest things I've seen in a really long time. That sounds really cool Roberto. Next Challenge, perhaps?

How it works (http://www.howstuffworks.com/augmented-reality.htm)

-Rage

[RobertoOrtiz]

It depends on two things
Interest (how many people are interested in this,
and some R&D I am doing on my end...

But keep in my that I have a degree in Comp Sci, so this is like me going back to my roots

[RageOfAges]

oh man that website is incredible. I work for a hotel (and do 3d as freelance). I emailed that website to my GM. I checked and our Hotel isn"t there, so I told him we should become a part of it.


-Rage

[razeverius]

Cool. Hopefully the challenge continues so I could see how its done :D

[WyattHarris]

Not sure how a challenge based on this would work but the tech is fascinating. I would definitely like to see more.

[Geta-Ve]

I would definitely be up for something like that!

[tuckerhcool]

I would be very interested in learning how to create models that work with this tech. I say if there enough people who understand the tech then go ahead with the challenge; for the techniques and knowledge that would come out of the challenge would be invaluable.

[RobertoOrtiz]

Lets use this thread, as and R&D thread for a future chalenge:


Well here are some cool demos...


link (http://createdigitalmotion.com/2009/01/07/happy-new-year-with-augmented-reality-flying-words-of-wisdom/)
Link (http://www.mikkoh.com/blog/?p=225)
Link (http://www.youtube.com/watch?v=xrsHBWk_z8Q)

And for those not afraid of some code:

Flash Augmented Reality
Getting Started (http://www.mikkoh.com/blog/?p=182)

[RobertoOrtiz]

oK HERE IS MRE



after checking this video
Papervision - Augmented Reality (extended) (http://vimeo.com/2283082) from Boffswana (http://vimeo.com/user946580) on Vimeo (http://vimeo.com/).

You can get thethe Flash movie AND The source files from this page (http://interactive.digitalpictures.com.au/?p=392).

[Boone]

So the camera uses a square piece of card, with identifiable shapes drawn on to it, to determine the model's position. That is actually a good idea, and seems to work very well. For a static model I would say the novelty would wear off quite soon, but with the addition of animations and vfx it looks fantastic.

Does this technique require a special camera, or am I correct that this only requires the software?

[menthol]

Cool, but for me, it's more Anim-Session/FX-Wars stuff than HCM. I'd rather we stick to the list (http://forums.cgsociety.org/showthread.php?f=208&t=711791).

[angel]

if it requires me to open Flash then count me out. Even tho the tech looks interesting and all I'm not really looking forward to learn this or Flash, heck I still need to learn MEL!

[axiomatic]

The performance of AR stuff is the only thing I'd worry about. I would have thought a challenge involving anything going into Papervision would really be quite difficult if only because of the triangle limits you'd need to stick within.

Now if we could make use of an Alchemy-based 3D engine that would perhaps make things more interesting :)

Anyways, if you need a hand with Flash or even just a couple of people to help porting models for an upcoming challenge I'd be happy to help. Have reasonably extensive AS2/3 experience.

[InKraBid]

Hey, I've been working with 2 developers on different occasions, one of them is my closest boss in my current job. Am very interested to hear about the project.

[RobertoOrtiz]

if it requires me to open Flash then count me out. Even tho the tech looks interesting and all I'm not really looking forward to learn this or Flash, heck I still need to learn MEL!
Actually no

The tech part (that is why I am doing the R&D as you read this) and it would be done WAY before the challenge starts.




The idea would be for the modelers to partner up with a coder (ie someone like me)

Let me research this more and if this works we would get a LOT of attention.

...trust me guys, I can make this work


Speaking or research,

BTW I welcome the fact that papervision has poly limits.

That makes the challenge even more cool

Here is more info on the topic:
Papervision_3D & Collada (http://www.collada.org/mediawiki/index.php/Papervision_3D)

And
Getting 3d models into Papervision) (http://www.mikkoh.com/blog/?p=227)


here are some cool Papervision tutorials for flash:
Introduction to Papervision (http://www.computerarts.co.uk/tutorials/3d__and__animation/introduction_to_papervision)
and

Source | Papervision3D FreeCamera Movement in AS3 (http://flashenabledblog.com/2008/05/16/source-papervision3d-freecamera-movement-in-as3/)

Papervision 3D Programming Tutorial - Loading and Displaying a 3D Model (http://www.brighthub.com/internet/web-development/articles/12902.aspx)

Papervision Programming Tutorial - Animations (http://www.bukisa.com/articles/32992_papervision-programming-tutorial-animations)

and
Papervision how to videos: (http://www.wonderhowto.com/search/papervision/)


Star (http://www.rockonflash.com/demos/pv3d/starwarsflightsim/index.html) Wars demo (http://www.rockonflash.com/demos/pv3d/starwarsflightsim/index.html)

-R (http://swut.net/q2lw.html)

[RageOfAges]

Here's a really cool one:

http://www.youtube.com/watch?v=kM6QDd0XqQ4&feature=related

-Rage

[Justame2002]

Could be interesting, looks a bit complex for people to participate though, more like matchmoving, nothing new really, you'd need some kind of real mockup, virtual content interaction, as well as Adobe Flash CS4 and actionscript coding skills

[axiomatic]

Could be interesting, looks a bit complex for people to participate though, more like matchmoving, nothing new really, you'd need some kind of real mockup, virtual content interaction, as well as Adobe Flash CS4 and actionscript coding skills

I would hazard a guess that Mr. Roberto is trying to get a simple workflow together so that effectively all you would need to do is do your modelling with a couple of extra rules (similar to v. low poly game modelling) and then upload it.

From there some server side code handles the model, checks it's valid etc, then you're given a dynamic URL for an image card ... (or perhaps you design the image card too - just for kicks).

Basically I imagine he's aiming for a solution where no one (except maybe a few AS bunnies) is going to need to touch Flash :)

[Boone]

ME! ME! I'M A CODER! PICK ME! MEEEEEEE! :cool:

That would be cool to team up with someone else for this...

[RobertoOrtiz]

I would hazard a guess that Mr. Roberto is trying to get a simple workflow together so that effectively all you would need to do is do your modelling with a couple of extra rules (similar to v. low poly game modelling) and then upload it.

From there some server side code handles the model, checks it's valid etc, then you're given a dynamic URL for an image card ... (or perhaps you design the image card too - just for kicks).

Basically I imagine he's aiming for a solution where no one (except maybe a few AS bunnies) is going to need to touch Flash :)
I could not have said it better...

:)

[grantmoore3d]

I would definitely be interested in that! I haven't touched Flash in a while, but I do enjoy playing with code for neat projects like this! I'll have to look into it next month when I've got some more time... any idea when you'd possibly hold the challenge?

[WyattHarris]

.:Raises hand:. Yes, I'm a coder too.

But another language?!? lol

[RobertoOrtiz]

I would definitely be interested in that! I haven't touched Flash in a while, but I do enjoy playing with code for neat projects like this! I'll have to look into it next month when I've got some more time... any idea when you'd possibly hold the challenge?
Coders you can start NOW...


The challenge will go live as soon as we can create a viable pipeline...
My intention is to do more R& D on this topic, and use this thread as a repository for my results. I want to encourage the other interested coders to do the same since there are enough tutorials on this thread to start doing some basic work.

As soon as we can publish standard collada models into this enviroment we will go live.

In other words want the modellers to only do models, and leave the coding heavy lifting to the flash developers.

[axiomatic]

Been doing some reading and poking and the following things might be handy to others who are also reading and poking...

Front end stuff:
This christmas AR app (http://specialmoves.com/xmas/) looks really promising as a starting point for what we need. It's a working AR flash app that has multiple 3D objects selectable via a menu.

Looking through the AS file and it seems that, sans-sound and extra interaction (i think the Doll has some on-click events), it would be very easy to parse an XML list of available models to flash player on the fly.

Also looks like you could easilly pull the pattern file location from XML and load a restricted set of models based upon that.


Back end stuff:
Saqoosha has an air app marker generator (http://saqoosha.net/lab/FLARToolKit/MarkerGenerator/MakerGenerator.air) that could possibly form the start of a means to generate individual markers for people participating and dump them into a usr folder etc.

There's additional information on ASCollada (handles the conversion of collada .dae into a PV usable object) in this post (http://modern-carpentry.com/talk/?p=30) on modern-carpentry.

I also found quite a lot of info on materials, shaders and textures for PV3D v2 on this blog (http://www.rustpunk.co.za/blog/index.php) by rustpunk.

In terms of referencing the DAE object this thread (http://www.nabble.com/Collada-file-component-referencing-td21636048.html) might proove useful for navigating the collada object. It also might yield some interesting concepts with regards to dynamically alterting textures etc.

That's all for now :)

[grantmoore3d]

Just realized I'm going to need to grab a webcam to be able to play with this... any suggestions (please post answers to this question on the following link) :

http://forums.cgsociety.org/showthread.php?f=23&t=740289

[axiomatic]

Just realized I'm going to need to grab a webcam to be able to play with this... any suggestions (please post answers to this question on the following link) :

http://forums.cgsociety.org/showthread.php?f=23&t=740289

Not sure about this ... but it might be nice to have a means to run an FLV that the AR is applied to as a quick means of testing and as a way for those without a camera to see how their model would look.

[RobertoOrtiz]

BigPixolin posted this on GD:


Webcam Brings 3-D to Topps Sports Cards

Since the 1950s, Topps has sold baseball trading cards filled with photos and stats, bringing the game to life. Now the company is bringing its cards to life.


http://www.nytimes.com/2009/03/09/t...ps.html?_r=1&em

[axiomatic]

You know what would be really cool is to somehow use the actual cards in a more dynamic way. I mean ... something like those table top miniatures games.

Or ... Battle Chess :D

Each piece could be an individual icon on the board and you play the game in front of you moving the cards. As a card gets overlaid by an existing card (collision etc) then an animation is triggered such as an old beat down :)

Not to mention practical purposes ... like automatically logging game moves etc.

[RobertoOrtiz]

You know what would be really cool is to somehow use the actual cards in a more dynamic way. I mean ... something like those table top miniatures games.

Or ... Battle Chess :D

Each piece could be an individual icon on the board and you play the game in front of you moving the cards. As a card gets overlaid by an existing card (collision etc) then an animation is triggered such as an old beat down :)

Not to mention practical purposes ... like automatically logging game moves etc.

That is a GREAT idea...

We could do a challenge called "Warriors of the Zodiac"

And people would have to model monsters based the symbols of the zodiac...
we could be inspired by the type of rules they use on miniature games like
heroclix...
http://www.heroclix.com/5th/

[axiomatic]

That is a GREAT idea...

We could do a challenge called "Warriors of the Zodiac"

And people would have to model monsters based the symbols of the zodiac...
we could be inspired by the type of rules they use on miniature games like
heroclix...
http://www.heroclix.com/5th/

Heroclix was awesome :)

Something like this has huge potential.

I think that's what's exciting about this tech ... there's a lot of ways to use it that aren't startlingly obvious from the get-go.

I think the ability to change the animation queues based on circumstances, or even just other simple sub-cards, will really send it rocking. You know, hold up your "Sonic Sting" card and your Scorpio executes it's special move animation.

Gah. Think I better sharpen up my AS3 skills and have more of a poke around. Just need to find the time :/

[RobertoOrtiz]

Well work hit me hard, but I am back at this..

here is a cool site with source files:
http://papervision2.com/

[axiomatic]

A little off topic:

I was thinking about augmented reality some more and other interesting ideas that could be done with it.

It would be very cool to make use of Papervision2's ability to plug other info into 3D objects. So for example, one could make an interface pop up in 3D from a card left on your desktop. Move the cards around physically and you could be moving 'windows' around in your augmented reality work space. Windows 3D :)

[grantmoore3d]

I've got a couple weeks left of school here so haven't had any time to try out some code just yet. However, I was poking around the web and found this :

http://graphics.cs.columbia.edu/projects/goblin/

Flash is much more marketable in the long run and would probably be better for the competition because it has poly limitations, but I think I'm going to look into this as a possible alternative. I'll post any results I get, though it will be a few weeks till I can really get started on the code.

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.