Researchers at Carnegie Mellon University have made tracking technology a common tech fixture in virtually every home.
They used radio signals from WiFi routers to detect and track the three-dimensional shape and movements of human bodies in a room.. They didn’t need to use cameras.
This is how the authors explain in their study:
We believe that WiFi signals can serve as a ubiquitous replacement for RGB images for human detection in certain cases. Illumination and occlusion hardly affect WiFi-based solutions used for indoor surveillance.
In addition, they protect people’s privacy and the necessary equipment can be purchased at a reasonable price. In fact, most homes in developed countries already have Wi-Fi in the home, and this technology can be extended to monitor the well-being of seniors or simply identify suspicious behavior at home.
The team used DensePose, a system for mapping all pixels on the surface of a human body in a photo, developed by researchers at Facebook’s AI Lab.
What makes DensePose powerful is its ability to identify more than two dozen key points and zones of the human bodysuch as joints and body parts such as arms, head and torso. This allows the AI to describe a person’s pose.
Combining all this data with a deep neural network, they were able to map the phase and amplitude of WiFi signals sent and received by routers to coordinates on human bodies.
For your demonstration, The researchers used three $30 Wi-Fi routers and three receivers in a row that bounce Wi-Fi signals off the walls of a room.
The system cancels out static objects and focuses on reflected signals from moving objects, reconstructing a person’s posture into a radar-like image. It was shown that it worked even if there was a wall between the routers and the subjects.
It’s not the first time researchers have tried to “see” people through walls. In 2013, an MIT team found a way to use cell phone signals for this purpose, and in 2018, another MIT team used WiFi to detect people in another room and translate their movements into puppets.
However, the Carnegie Mellon team’s new study offers much higher spatial resolution. You can see what moving people are doing by looking at their poses.
The researchers believe thatWiFi signals “could serve as a ubiquitous replacement” for regular RGB cameras, and cite several advantages such as the ubiquity of these devices, their low cost, and the fact that using WiFi overcomes obstacles such as poor lighting and occlusion that normal camera lenses face. They add that they can be detected and pointed “suspicious behavior” inside a house.
