After many hours of busywork and fiddling around with code, I have finished writing an OpenCV template. The main point of it is to create a platform that initialises the image grabber in a separate thread. This can be a feat for many beginners, like me a couple months ago. To make the transition into high performance Vision Applications with OpenCV, I have written this. It theoretically supports the Kinect, but I will test that tomorrow when I have access to a Kinect.
When you first open the files, you may realise that there are more comments than there are lines of code 
:yikes: . I have written this to help teach students C++, because the student I am teaching knows really the basics of C++, the Hello World! application!
Even though most of the comments look stupidly beginner level, read them because they include a TON of information on how to get your platform set up.
A nice thing is that if your code works on this in one computer, it will work identically in another system, almost always.
Now that my time is opening up, I’d be happy to help a couple of people learn at least the basics of AI! I remember that it was quite hard to get started with OpenCV because many of the tutorials out on the web are for OpenCV 1.x, and I wanted to learn the more simpler (and more powerful) C++ API. The online documentation for many mediocre-complex functions is not so human-understandable!
The template is available here
If you do not have OpenCV + libfreenect + libfreenect-sync installed under Ubuntu, I have finished my platform installer. Download the main script from here
Please feel free to PM me or reply to this post if you have any questions, suggestions, or anything else! 
Good luck and hopefully, your summer break is right around the corner!