With the help of AI neural networks and deep learning, researchers at Carnegie Mellon University were also able to create full-body images of subjects. This proof-of-concept would be a breakthrough for healthcare, security, gaming (VR), and a host of other industries. It would also overcome issues affecting regular cameras, such as poor lighting or simple obstacles like furniture blocking a camera lens, while also eclipsing traditional RBG sensors, LiDAR, and radar technology. It also would cost far less and consume less energy than the latter, researchers noted. However, this discovery comes with a host of potential privacy issues. If the technology does make it to the mainstream, one’s movements and poses could be monitored — even through walls — without prior knowledge or consent.
Perceiving Humans Via WiFi Antenna, Seeing Past Obstacles
Researchers used three WiFi transmitters, such as those on a $50 TP-Link Archer A7 AC1750 WiFi router, positioned it in a room with several people, and successfully came up with wireframe images of those detected in the room. With the help of artificial intelligence algorithms, researchers managed to create 3D images from the WiFi signals that bounce off of people. Technically speaking, researchers analyzed the amplitude and phase of the WiFi signal to find human ‘interference’ signals and then later allowed AI algorithms to produce an image. “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 researchers’ pre-print brief reads. DensePose itself is a technology developed by Meta Platforms Inc. and used by Facebook that generates 3D visuals of humans via a flat RGB image, not via WiFi radio signals as is the case here. The research paper goes on to say that WiFi perception will quell privacy concerns that arise when traditional sensors such as cameras are used in non-public places. “We believe that WiFi signals can serve as a ubiquitous substitute for RGB images for human sensing in certain instances,” researchers said. “In addition, they protect individuals’ privacy and the required equipment can be bought at a reasonable price. In fact, most households in developed countries already have WiFi at home, and this technology may be scaled to monitor the well-being of elder people or just identify suspicious behaviors at home.” Researchers noted some of the shortcomings of the technique, such as unusual body poses and difficulty with three or more subjects in the space that can be resolved by obtaining more training data for AI algorithms. “In future work, we also plan to collect multi-layout data and extend our work to predict 3D human body shapes from WiFi signals. We believe that the advanced capability of dense perception could empower the WiFi device as a privacy-friendly, illumination-invariant, and cheap human sensor compared to RBG cameras and Lidars.”
Potential Privacy Dilemma on the Horizon
Although researchers said the technology could be used for good such as monitoring the well-being of elderly people without disturbing them with cameras or detecting break-ins by thieves, there are serious privacy issues that may arise if the technology were to become mainstream. In a time where facial recognition, doorbell cameras, drones, and hackable IoT devices such as indoor smart cameras are abundant and endanger our privacy and security every day, WiFi perception technology would be the cherry on top of everything. It could be abused by everyone from Big Data to cyber criminals and could make people lose trust in their humble WiFi routers. Detecting people without cameras or expensive LiDAR (Light Detection and Ranging) sensors is not new. In 2013, researchers at MIT found a way to use mobile phone signals to breach walls, and in 2018 another MIT team came up with a more basic version of the technique discussed above. If you have an iPhone 12 Pro or later, or the 2020 iPad Pro or later, they come with a LiDAR sensor, which is a pulsed laser beam, mainly for AR (augmented reality) applications. You can try out what 3D object mapping looks like by downloading the free Polycam app on the app store. This type of bleeding-edge technology, one that may potentially see through walls, is reminiscent of a scene from The Dark Knight film. It may well one day replace cameras and other sensors and may be part of the smart cities we shall soon live in.
