Before I go on, I would like to apologize if this website isn't the best place to ask this question. As far as the stack exchange network goes, I believe this is the most applicable place for this.

As a computer engineer undergraduate, I'm planning on working on a graduation project that involves the growing concept of augmented reality. This will be a long-term project, and I would really like to work on something that, ideally, would put me at the top of the field once I'm done. I'll have a full year to work on this, and will have all the help and resources I might need, be it from my university, from the company i'm interning in next summer, or whatever.

The problem is, since this concept is still in its infancy, I don't really know where to start, or what to accomplish. There are some good frameworks developed and ready to be used on personal computers or mobiles, but I'm really at a loss of what to do with them. I even went as far as trying to develop a simple framework of my own, but I'm not really sure if this is anything practical, or even doable with my current knowledge and skills.

So I guess my question is, is work in this field something that I could / should delve in? Do you guys have any suggestions that can help me, or am I better off with something entirely different? Is there really any practicality in anything I could do at this point in time, or should I just wait for the giants (like google) to make more progress in this field before I, as an undergraduate student, work on it?

I'm really at a loss here, and hopefully I made myself clear enough.

  • Hi, @Malfunction, and welcome! Do you have to actually build it, or can you conceptualize it and build a proof of concept? Feb 7, 2014 at 19:20
  • I would probably need to make something tangible, and even if I don't, I would like to. So yeah, something to be built! Feb 7, 2014 at 19:28
  • I think you're in a great position to do something interesting. Look for pain points with the current state of augmented reality, and focus on the one that intrigues you, then talk with your professors about the viability of your idea. At least, that's what I would do. Feb 7, 2014 at 20:27
  • @Malfunction How is your understanding of computer science? You've got some complicated problems to solve before you can create a usable interface. The Kinect makes your job MUCH easier than it used to be though. Feb 7, 2014 at 21:34
  • 1
    @LindaBrammer That's actually something I haven't thought of. Thanks for your input! Feb 8, 2014 at 6:59

1 Answer 1


It sounds like your question is: what possible practical applications are there given the current state of augmented reality technology. Is that fair?

I don't really know the current state of augmented reality technology, but the two actually useful AR implementations I've seen are:

  1. Augmented Reality for Task Localization in Maintenance of an Armored Personnel Carrier Turret: Military, head-mounted repair assist.

  2. Word Lens mobile app: translate and replace text.

Both of these are useful examples because they limit the scope of the problem. Neither try to do AR over arbitrary imagery, but rather choose a relatively small slice of the world to deal with, the confines of a tank and printed words, respectively. This makes their task much much easier, and, in fact, possible given the state of the technology.

So my question to you, as you consider the current state of augmented reality is: what can you recognize or map? Can you limit your scope in a way that allows you to make the most out of the tech?

And, alright, here's my idea for you: make a real-life room-escape game (see Real Escape Game for what I'm talking about) that utilizes phone-based AR. You will know everything in the room so your image recognition problem gets a lot simpler. The phone could act as a virtual clue system, giving background on different objects that could move the game forward.

  • Limiting the scope of the problem is indeed the best approach to such a big undertaking. Thank you very much for your response. Feb 18, 2014 at 20:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.