Imagine being able to see through walls and identify who stands on the other side.
A new technology called RF Capture brings us one step closer to “x-ray vision.” The technology picks up wireless reflections off the human body in order to “see” the silhouette of a person standing behind a wall.
The test device is installed on one side of a wall. The device then transmits wireless signals that can traverse the wall and reflect off a person’s body back to the device.
The receiver captures these reflections and analyzes them to develop the image of a person’s silhouette.
But the team at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) who designed and built this technology didn’t just want to see if a person was standing behind a wall. It wanted to be able to identify different individuals and track movement.
How the RF Capture Works
The challenge here is that different people – and different body parts – all reflect the same signal.
The team had to devise a method to analyze the reflections in a way that could differentiate between different limbs and different people.
“The data you get back from these reflections are very minimal,” says Dina Katabi, director of the Wireless@MIT center. “However, we can extract meaningful signals through a series of algorithms we developed that minimize the random noise produced by the reflections.”
By testing different human subjects and using metrics such as height and body shape to create “silhouette fingerprints,” researchers can use wireless reflections to differentiate between individuals on the other side of a wall. (Image courtesy of Fadel Adib/CSAIL.)
The technology operates in two stages. To start, the device scans 3D space to capture the wireless reflection off objects in the environment, including the human body.
However, since only a subset of body parts reflects the signal back at any given point in time, the device then monitors how these reflections vary as someone moves through the environment.
Software then intelligently stitches the person’s reflections across time to reconstruct the silhouette into a single image.
To differentiate between people, the team then repeatedly tested and trained the device on different subjects. Using metrics such as height and body shape the device can create “silhouette fingerprints” for each person.
Safety, Search and Rescue and the Smart Home
The CSAIL’s wireless imaging technology is also able to track human movement accurate enough to determine if someone is breathing.
This means the RF Capture technology could have significant advantages in risky operations such as firefighting and search and rescue.
Being able to “see” through walls or debris and identify not only if there is a human body on the other side but whether that body is breathing could mean more efficient rescue operations and greater safety for rescue workers.
The team also sees safety benefits in the home, especially when integrated into the “smart home”.
They suggest that future versions of this technology could monitor the occupants of a home and be programmed to call emergency services if the system identifies that someone has fallen down or become unconscious. This would be very useful for caregivers of elderly or ill family members.
There are also the more mundane operations within a smart home environment, such as monitoring the occupants and adjusting lights, temperature or operating appliances based on who is located in which room.
Film and Videogame Applications
More creative applications include improved motion capture for filmmaking and videogame interaction.
“Today, actors have to wear markers on their bodies and move in a specific room full of cameras,” says PhD student Fadel Adib. “RF Capture would enable motion capture without body sensors and could track actors’ movements even if they are behind furniture or walls.”
Future versions could also be integrated into gaming interfaces, allowing players to interact with a game from different rooms or even trigger distinct actions based on which hand they move or how they stand.
The team compared the gaming capabilities of its device with the Kinect. The team found a high correlation between the motion tracking of both devices. However, the Kinect needed to be pointed at the individual, be in the same room and have an unobstructed field of view.
“The possibilities are vast,” says Adib. “We’re just at the beginning of thinking about the different ways to use these technologies.”
However, as with any new technology with the capability to locate, image or identify individuals, privacy is always a concern. Users will want to know what methods will be available to protect them from unwanted observation.
For more on the RF Capture technology, check out the team’s paper or visit the project website.