MIT researchers devise simple, inexpensive gesture-based computing system
Researchers at the Massachusetts Institute of Technology have developed a system to make gesture-based computing more natural and less expensive.
Instead of arrays of multiple cameras or sensors built into screens, MIT's solution involves a standard webcam and a multicolored Lycra glove that costs $1 to manufacture.
Developed by graduate student Robert Wang and professor Jovan Popović, the glove eschews the current two-dimensional method, which uses tape placed on fingertips, for a three dimensional approach that can capture flexing fingers with almost no lag time.
The gloves have 20 irregularly shaped patches in 10 different bright colors, allowing the system to better distinguish between them. The arrangement and shapes of the patches allow the system to tell the front and back of the hand apart.
On the computing side, a new algorithm helps the computer quickly call on visual data stored in a database. Once a webcam has captured an image of the glove, Wang’s software crops out the background, superimposes it on a white background, reduces the resolution of the cropped image to just 40 by 40 pixels and seeks a pair in a database containing 40x40 hand models in various positions.
Since the system doesn't have to calculate the relative positions of the fingers, palm, and back of the hand on the fly, it can respond in a fraction of a second.
The system must first be calibrated using a standard sheet of paper on a flat surface in front of the webcam. The process takes about three seconds.
The system's simplicity offers an inexpensive solution to immersive video games and, by extension, engineering and design projects that require the manipulation of 3D models.
What's more, Wang is exploring how to extend the system to capture full-body motion, using shirts and pants.
Here's a look at the system in a video:
The project was first unveiled at last year's Siggraph computer graphics conference.
This post was originally published on Smartplanet.com