Most Read Latest News Blog Resources
Digg!  Digg
Reddit  Reddit


            iphoneapp GET THE APP!

Windows & .NET Watch: Prototyping with Processing




July 15, 2009 —  (Page 1 of 3)
Recently, I had a client who was interested in “augmented reality” applications. The phrase is used to describe applications that overlay computer-generated imagery on a heads-up display or, at least, a live video feed. This type of thing is old hat at ACM SIGGraph but still pretty radical for a consumer website marketing campaign.

Perhaps one day I’ll have one of those clients that says, “Take your time and let us know what you discover. We’re writing this development effort off as high-risk, high-payback research and development.” They’ll pay in pixie dust and give me rides on their flying horses. Until then, like most post-collegiate developers, I have to be careful about embarking on “interesting” programming projects.

In the case of augmented reality, success depends on recognizing targets within video frames. The targets take up relatively few pixels and have lower contrast and far more noise than in a reference photo. Even a plain vanilla video display eats up a significant amount of local resources, and in addition to finding the targets, you presumably have to, you know, do something with the information, such as reconstruct the spatial location of the targets, project that onto some model, and composite that into the output. And then you have to do it all over again in the next several dozen milliseconds.

If you are interested in video processing, you should probably begin with OpenCV and, if possible, combine it with Intel’s Integrated Performance Primitive libraries. Unfortunately, for various reasons this was inappropriate for my client’s application. Strangely, this actually lowered my risk profile for the project; had I the advantages of object-recognizing libraries and SIMD instructions, I definitely would have faced the risk of not properly taking advantage of them!

Instead, I had a straightforward brick wall to get through: Could I reliably do target detection in a video frame? Further, I knew that the client would not fund the development of a sophisticated object-detection algorithm (no Viola-Jones boosted rejection cascades for me!). If I couldn’t achieve the detection goals with a relatively simple algorithm, I would have too little traction on the problem to commit to a schedule for the whole project.

Related Search Term(s): Processing

Pages 1 2 3 


Share this link: http://www.sdtimes.com/link/33614
 

Add comment


Name*
Email*  
Country     


  • Comment
  • Preview
Loading



 
 
 
 
News on Monday
more>>
SharePoint Tech Report
more>>


   

 
 
Download Current Issue
ISSUE 9/1/2010 PDF

Need Back Issues?
DOWNLOAD HERE

Receive the print Edition?


 
blogs tab
VMworld hops to it
Data center operating systems play a big part at VMworld, but it's still too soon.
09/02/2010 01:42 PM EST

Certificate program for secure cloud computing
The Cloud Security Alliance introduces user certification.
09/01/2010 04:20 PM EST

What does the Army's Crusher tank and RIM's tablet computer have in common?
RIM plans to use Crusher tank technology on its yet-to-be-announced tablet.
08/25/2010 04:16 PM EST

 

Events calendar tab
9/13/2010 to 9/15/2010
San Francisco
Intel

9/19/2010 to 9/23/2010
San Francisco
Oracle

9/19/2010 to 9/23/2010
San Francisco
Oracle

9/20/2010 to 9/23/2010
Boston
TechInsights

9/20/2010
New York City
Flagg Management