Mapping Yourself and Your Gaming Room

A while back, I discovered that a method for increasing pointer accuracy used in the setting screens for the Wii version of Zelda: Twilight Princess could be used to tell the game where your screen is in 3D space relative to the sensor bar. Then, if the system knew where the sensor bar was in 3D space relative to the Wii remote, it would know where the screen is relative to the Wii remote.

However, I've just realized there's a much easier way to tell the system where your TV is in 3D space.

All you need to do is touch at least three corners of the TV with the tip of a Wii MotionPlus-equipped Wii remote. (WiiMot+) The accelerometer in the WiiMot+ could even be used to automatically detect when it touches an object.

This method can be used to map out your entire gaming room virtually. For example, once you map out the location of your screen, you could tell the system how low the floor is relative to the screen by simply placing the WiiMot+ on the ground. Any stationary obstacles in your gaming room (such as a couch or standing speakers) could be mapped out in the same way.

NOTE: Once the system knows where the screen is in 3D space relative to the WiiMot+, the system becomes capable of knowing when the WiiMot+ is pointing at the screen and where, without using the sensor bar. (although the sensor bar would probably still be useful for verifying the pointer's location)

All this virtual mapping may be neat to think about, but this information isn't particularly useful to the system until you also tell it where you are in 3D space. That is, until you give the system a "Body Frame Approximation". Once it knows this, it's possible for it to render your entire body virtually and track it in real time. Useful proportions might include such measurements as: how high your waist is above ground, the height of your knees above ground, the width of your shoulders, and the distance between your shoulder and your elbow. All of these measurements can be mapped in the same way the TV screen was.

There's a problem though. Even if the system knows where the hand holding a single WiiMot+ is in 3D space relative to the screen and/or the floor, it can't make any guesses where the rest of your body is. You could potentially move your body in any direction while keeping the in-hand WiiMot+ stationary. The system would need to know the locations of all relevant "independent appendages" (e.g. both hands) and one "fixed appendage." (e.g. your head) Luckily for us, "3IRLED head tracking" and "twomote" fulfill these requirements perfectly.

NOTE: In order for the tracking to be accurate, each WiiMot+ must be calibrated to an initial position like I described in my last post before any other objects or body proportions are mapped. This is only necessary if head tracking or twomote are used. With head tracking, the system needs to know the head tacker's location relative to the TV. Likewise, with "twomote" the second WiiMot+ needs to know where it is relative to the mapping done with the first WiiMot+. Technically, one way to get around this problem could be to map the second WiiMot+ as an object using the first WiiMot+.

NOTE: Technically, you could move your elbows and legs around to a limited degree while keeping the two in-hand Wii MotionPlus-equipped Wii remotes and your head stationary, but not enough to affect gameplay.

How's that for an immersive experience?