Yes, scientists have developed technology that allows them to use WiFi signals to detect and track people’s body positions inside buildings. Researchers at Carnegie Mellon University have developed a method for detecting the three-dimensional shape and movements of human bodies in a room using only WiFi routers.
DensePose
They achieved this by using DensePose, a system for mapping all the pixels on the surface of a human body in a photo, and a deep neural network that maps WiFi signals’ phase and amplitude to coordinates on human bodies[1][2]. This technology has the potential for various applications, including monitoring the well-being of elderly people or identifying suspicious behaviors at home[2]. However, it also raises concerns about privacy and security, as external attackers could potentially use this technology to monitor activity inside a building without being detected themselves[3].
Countermeasures
None have been discussed in the research at this time. The potential for misuse of DensePose for surveillance or law enforcement has been highlighted, but specific countermeasures or ways to thwart its use are not discussed in the available sources. Therefore, the methods to thwart DensePose technology are not explicitly outlined in the provided information.
Implementation
The implementation of DensePose technology requires the following:
1. Training Data: Annotated training data is needed to train the deep-learning system. Facebook researchers initially enlisted human annotators to create a training dataset by manually labeling certain points on images. The task of labeling was broken down into body segments such as head, torso, limbs, hands, and feet to improve accuracy[6].
2. Algorithm and Model: DensePose technology is based on a deep-learning system that can transform 2D photo and video images of people into 3D mesh models of their bodies by estimating the positions of their torsos and limbs. The system uses an algorithm to estimate and fill in the points that correspond between 2D images[6].
3. Hardware and Performance: The system is capable of performing the 2D to 3D conversion at a rate of 20-26 frames per second for a 240 × 320 image, making it generally capable of creating 3D models of humans in a 2D video in real time[6].
4. Code and Software: Facebook has publicly shared the code for its DensePose technology on the software development platform GitHub, which could be used by developers and researchers for various applications, including graphics, augmented reality, or human-computer interaction[6].
While the specific technical details of the implementation are not provided in the search results, the information highlights the need for annotated training data, the use of deep-learning algorithms, and the availability of the code for DensePose technology on GitHub for implementation purposes.
Facebook, the Anti-Privacy Company
Facebook, often associated with privacy concerns, has been involved in various privacy issues and scandals. These include instances where individuals’ identities and private information were compromised without their permission, as well as the unauthorized collection of personally identifiable information of millions of users. The company has faced criticism for its handling of user data, including instances where it falsely claimed that third-party apps could access only the data they needed to operate, while in reality, these apps could access nearly all of a user’s personal data.
Facebook’s history of privacy issues has led to regulatory actions, settlements, and fines, including a $5 billion penalty imposed by the Federal Trade Commission (FTC) in 2019, which also included sweeping new privacy restrictions on the company[11][12][13][14][15]. These incidents have contributed to ongoing concerns about the protection of personal data on Facebook and its associated platforms.
DensePose Facebook Connection
Facebook’s involvement in DensePose technology is primarily through its research arm, Facebook AI Research (FAIR). In early 2018, Facebook’s AI researchers unveiled a deep-learning system that can transform 2D photo and video images of people into 3D mesh models of their bodies.
The technology, named DensePose, goes beyond basic object recognition, as it can make 3D models of human bodies by estimating the positions of their torsos and limbs. While Facebook has highlighted potential applications of DensePose in graphics, augmented reality, and human-computer interaction, concerns have been raised about the troubling implications of this research, particularly its potential for real-time surveillance.
Facebook publicly shared the code for its DensePose technology on the software development platform GitHub, making it accessible to developers and researchers. The release of this technology has raised concerns about its potential misuse for surveillance or law enforcement purposes, as it could be adapted by others for such applications. However, Facebook’s researchers did not specifically mention surveillance as a possible application of DensePose alongside the many other potential uses[6].
Citations
[1] https://www.popularmechanics.com/technology/security/a42575068/scientists-use-wifi-to-see-through-walls/
[2] https://www.vice.com/en/article/y3p7xj/scientists-are-getting-eerily-good-at-using-wifi-to-see-people-through-walls-in-detail
[3] https://news.uchicago.edu/story/how-hackers-could-use-wi-fi-track-you-inside-your-home
[4] https://science.slashdot.org/story/23/01/18/1841207/scientists-are-getting-eerily-good-at-using-wifi-to-see-people-through-walls-in-detail
[5] https://www.reddit.com/r/AR_MR_XR/comments/1096mm0/full_body_tracking_with_wifi_signals_by_utilizing/
[6] https://spectrum.ieee.org/surveillance-concerns-follow-facebooks-densepose-tech
[7] https://arxiv.org/pdf/2110.15267.pdf
[8] https://github.com/facebookresearch/DensePose/security
[9] https://github.com/facebookresearch/DensePose/blob/main/INSTALL.md
[10] https://www.robots.ox.ac.uk/~vedaldi/assets/pubs/shapovalov21densepose.pdf
[11] https://en.wikipedia.org/wiki/Privacy_concerns_with_Facebook
[12] https://www.nbcnews.com/tech/social-media/timeline-facebook-s-privacy-issues-its-responses-n859651
[13] https://www.techrepublic.com/article/facebook-data-privacy-scandal-a-cheat-sheet/
[14] https://terranovasecurity.com/blog/data-privacy-scandal-facebook/
[15] https://www.ftc.gov/news-events/news/press-releases/2019/07/ftc-imposes-5-billion-penalty-sweeping-new-privacy-restrictions-facebook