Viability of wiimote project

Nov 29, 2010 at 3:54 PM

As I look at the latest commercially available emerging technologies (Kinect for example), I have to ask myself if the potential of the wiimote as an innovative pc interface device is coming to an end.  I am interested in hearing what others think.  For example, does the wiimote have any advantages over the Kinect for 3D position sensing?  What is its future?

Coordinator
Nov 29, 2010 at 9:01 PM
RussAnderson wrote:

As I look at the latest commercially available emerging technologies (Kinect for example), I have to ask myself if the potential of the wiimote as an innovative pc interface device is coming to an end.  I am interested in hearing what others think.  For example, does the wiimote have any advantages over the Kinect for 3D position sensing?  What is its future?

Two in my opinion...

- cost

- processing of data...it takes a lot of cycles to process video/depth coming from a Kinect

That said, you can do a whoooooole lot more with a Kinect than you can a Wiimote....  I think they'll both have their place in the short term, but it's likely the Kinect will take over.

Dec 13, 2010 at 2:52 PM

I've worked with the Wiimote over the past couple of months, and I've been watching the Kinect make it's debut..

The Wiimote has never truly been a reliable way of 3D positional sensing, since all you can calculate is relative position. In cases where that is what you want, then it is good. If you want absolute positioning, you'll have to go to some other technology.

http://www.sixense.com/
This product, for example, is essentially a Wiimote but with the Absolute positioning.

Ultimately, it comes to what are you trying to detect: If you are trying to calculate reliable, absolute position, the Wiimote has never been the right choice. That being said, the Wiimote does generally win for what it was designed for: Short movements like sword swings, and pointable based movement. The Kinect will dominate what it is designed for: Tracking people and 3D modelling.

And in reply to peekb, ideally, cost should never be considered unless there is a drastic difference between two products. Sure, the Wiimote is $50 while the Kinect is $150, but you only need to buy one Kinect rather than 4 Wiimotes (or 5 if you throw one through the TV). What you should really focus on is what is the best, most reliable method for tracking what you need to track. If you got a sword or gun game method, or you can take advantage of the Wiimote's IR Tracker, it should be used. Otherwise, use something else.

Jun 30, 2011 at 11:59 PM

After spending the last two months experimenting with different input devices, I'd almost always choose a Wiimote over a Kinect for the kinds of interfaces we develop.  Kinect has an effective absolute pointing resolution of about 10x10, on the XBox games that use the Kinect have three or four controls on a typical screen.  Due to the lack of a controller the Kinect also requires all of the other people in the area to be well behaved, such as all standing to the right or left, or well behind the person controlling the application, dwell to click slows down interaction, and holding your arms out and still for an extended period of time is difficult, while moving while the Kinect is enabled risks causing some unintended action.

The Wiimote has an effective absolute pointing resolution of about 500x500.  Still not enough to substitute as a mouse pointer, but closer to adequate.  More effective IMO is to use the motion plus to integrate Yaw and Pitch and create a relative pointing device.  Resolution becomes effectively unlimited and the person controlling the application no longer needs to be facing the screen (or emitter or whatever) to control it.  The buttons and the device help as well by making it clear which user is interacting with the machine -- passing control is the same as passing an object, simple and intuitive.

Jul 7, 2011 at 6:52 AM

My personal project right now uses a Kinect with a Wiimote; the Wiimote gives low-latency finger gesturing while the Kinect handles positional input.  Should be fun!  So I owe a double debt to the redoubtable Mr. Peek, because he is not only the author of WiimoteLib, but also involved in developing some of the Kinect SDK samples :-)

Cheers,

Rob