Microsoft's Kinect Allows Researchers To Improve 3D Scanning By 1,000 Times


Microsoft's Kinect is useful for more than just gaming.

In fact, many would argue that the device is better for things other than gaming. Even Microsoft has stopped including Kinect in its Xbox One bundles.

A group of researchers, however, is using the Kinect for 3D mapping. While that in itself isn't all that interesting, the researchers are looking to change the game and have developed a new system that offers a massive 1,000-times-better mapping than other software.

"Today, they can miniaturize 3D cameras to fit on cellphones," said Achuta Kadambi, a Ph.D. student at the MIT Media Lab and one of the developers behind the project, in a statement. "But they make compromises to the 3D sensing, leading to very coarse recovery of geometry. That's a natural application for polarization, because you can still use a low-quality sensor, and adding a polarizing filter gives you something that's better than many machine-shop laser scanners."

The researchers are from MIT, and their new system is called Polarized 3D, which essentially uses polarized light to map a surface in 3D. Of course, it is possible to calculate 3D models, however, in the resolution that the team is aiming for, massive amounts of raw data would need to be computed, making it a non-viable option. That's where Kinect comes in.

The Kinect is really good at measuring the depth of objects that it can see, and when the researchers combined the information from polarized light with the depth information from the Kinect, they got a very detailed image in 3D.

The researchers then used a Kinect camera with a polarizing photographic lens in front of it, and after processing the images with their algorithm, they achieved some pretty spectacular 3D renderings. In fact, when they put their renderings against a laser scanner, the renderings from the Kinect were far more accurate.

Of course, it will probably be a while before the new system finds any commercial use, however, the team says it is planning on releasing its source code, meaning that anyone can use the Kinect to process 3D images.

Via: Slashgear

ⓒ 2018 All rights reserved. Do not reproduce without permission.
Real Time Analytics