Sign In

Communications of the ACM

ACM TechNews

How Wireless 'x-Ray Vision' Could Power Virtual Reality, Smart Homes, and Hollywood

View as: Print Mobile App Share:
Researchers can use wireless reflections to identify from the other side of wall.

The new RF Capture technology can identify wireless reflections from the human body and use them to visualize the silhouette of a person behind a wall.

Credit: Fadel Adib/CSAIL

The new RF Capture technology developed at the Massachusetts Institute of Technology's (MIT) Computer Science and Artificial Intelligence Lab (CSAIL) can pick up wireless reflections off the human body to visualize the silhouette of a person concealed behind a wall.

Tracking the silhouette enables the device to trace the person's moving hand as well as differentiate between 15 different people through a wall with almost 90-percent accuracy.

RF Capture beams wireless signals that traverse the wall and bounce off a person's body back to the device, and these reflections are captured and analyzed to see the person's silhouette. RF Capture scans three-dimensional (3D) space to capture wireless reflections off objects in the environment, and then monitors how the reflections vary as someone moves in the space and stitches the person's reflections across time to rebuild his silhouette into a single image. To distinguish between people, the CSAIL researchers tested and trained the device on different subjects, using various metrics.

They envision the technology having significant ramifications for many industries. In filmmaking, RF Capture could facilitate motion capture without body sensors, according to the researchers.

MIT professor (and 2012 Grace Murray Hopper Award co-recipient) Dina Katabi says potential smart-home applications could include automatic 911 alerts when a resident has fallen unconscious, or intelligent in-home systems operation.

From MIT News
View Full Article


Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account