A team at Carnegie Mellon University has developed a technique of tracking human bodies through walls using WiFi signals. While the technology has already existed for a few years, the new technique has the WiFi signals sending and receiving a body’s coordinates and then using DensePose to map the body. DensePose maps human pixels in an RGB image to the 3D surface of the human body.
The research team states that this technique paves the way for low-cost, broadly accessible, and privacy-preserving algorithms for human sensing. Future plans include collecting multi-layout data and extending their work to 3D human body shape prediction via WiFi signals.
As stated in their paper: “We developed a deep neural network that maps the phase and amplitude of WiFi signals to UV coordinates within 24 human regions. The results of the study reveal that our model can estimate the dense pose of multiple subjects, with comparable performance to image-based approaches, by utilizing WiFi signals as the only input.”
The new technology is of particular use in low-light environments, with occlusion, and multiple people.
“Advances in computer vision and machine learning techniques have led to significant development in 2D and 3D human pose estimation from RGB cameras, LiDAR, and radars. However, human pose estimation from images is adversely affected by occlusion and lighting, which are common in many scenarios of interest.”
By reducing the need for expensive and advanced technology, the researchers say they can make human tracking more available. They claim that the breakthrough is a privacy-positive situation, with practical applications in monitoring the well-being of elderly people or identifying suspicious behaviors in the home. Others, however, have expressed concerns regarding its potential usage and privacy.