WiFi signals can map body movement through walls, but the biggest claims still need proof

A fast-growing open-source project called π RuView says it can turn WiFi signal data into human pose estimation, presence detection, and even breathing or heart-rate monitoring without a camera. The idea is real at the research level, and it builds on published Carnegie Mellon work. But some of the strongest claims now circulating online come from the project’s own documentation, not from independent testing or a peer-reviewed product validation.

That distinction matters. People should take this seriously because WiFi sensing has moved far beyond theory. Carnegie Mellon researchers already showed that a deep neural network could map WiFi phase and amplitude data to dense human pose regions, with results they described as comparable in visual performance to image-based approaches. At the same time, the jump from lab research to real-world, through-wall, low-cost deployment needs more scrutiny than many viral posts suggest.

RuView, published on GitHub by Reuven Cohen, presents itself as a privacy-first system for real-time pose estimation from WiFi Channel State Information, or CSI. The repo says the full feature set requires CSI-capable hardware such as ESP32-S3 or research-grade network hardware. It also says ordinary consumer WiFi gear does not expose the needed CSI data, which weakens the idea that this works out of the box on any random home router.

The public concern is still valid. If a system can infer movement, posture, and possibly vital signs from radio behavior instead of a camera feed, then people may not notice it at all. European privacy regulators have already warned that new tracking techniques and access to information from a user’s terminal equipment can trigger legal concerns, especially when identifiers or device-side data get used for tracking.

What is π RuView?

π RuView is an open-source GitHub project that claims to turn WiFi CSI data into:

• human pose estimation
• presence detection
• vital-sign monitoring
• multi-person tracking
• through-wall sensing

The repo also advertises fast processing, Docker-based setup, ESP32 support, and a live visualization stack. GitHub showed the project at about 31.5K stars when checked, which helps explain why it spread so quickly this week.

What the underlying research actually shows

The strongest independent foundation comes from Carnegie Mellon University and the related arXiv paper, “DensePose From WiFi.” In that work, researchers said they trained a neural network to map WiFi signal phase and amplitude to UV coordinates within 24 human body regions. They said the system could estimate dense pose for multiple people using WiFi signals alone.

That means the core concept is not science fiction. WiFi-based human sensing has credible academic backing. It also means the internet headlines are not entirely wrong when they say WiFi can “see” people in a room. But the academic paper does not automatically prove that every current GitHub implementation achieves the same real-world accuracy, speed, range, and reliability across ordinary buildings and noisy environments.

Where the sample article goes too far

The sample article treats several project-level claims as settled fact. That is too aggressive.

Here is the safer reading of the evidence:

Claim	What public sources support	Confidence
WiFi can support dense human pose estimation	Supported by CMU research and arXiv paper	High
RuView is a real open-source project with active code and documentation	Supported by GitHub repository and release history	High
Full sensing requires special CSI-capable hardware	Explicitly stated in the repo	High
Ordinary WiFi infrastructure alone can do all of this with no special access	Not supported by the repo; repo says CSI-capable hardware is required	Low
RuView delivers every advertised metric in real-world settings	Mostly comes from the project’s own readme, not independent validation	Medium to low
Privacy and surveillance concerns are real	Supported by the nature of the tech and EU tracking guidance	High

Why security teams should care

Security teams do not need to accept every viral claim to treat this as important. The bigger issue is that radio-based sensing can infer human activity without a visible camera, and users may never realize someone deployed it nearby. That alone creates a very different risk profile from traditional video surveillance.

The repo says the system can run with ESP32-S3 hardware, stream data, and feed a sensing server for visualization and analytics. Even if some performance claims turn out optimistic, the barrier to experimentation has clearly dropped. Cheap parts, open code, and online tutorials often push emerging sensing methods into broader use before policy catches up.

What this means in practice

For most readers, the key takeaway is simple:

• The research is real.
• The open-source implementation is real.
• The privacy risk is real.
• Some of the headline-level deployment claims still need independent proof.

That is a stronger and more accurate story than saying the whole thing is either fake hype or fully production-ready. The truth sits in the middle.

What to watch next

• Independent lab tests of RuView’s claimed accuracy, range, and reliability
• Whether hardware vendors lock down or expose CSI access more widely
• Whether regulators update privacy rules to address radio-based sensing
• Whether enterprise security teams start treating RF sensing as a physical security issue, not just a networking issue

FAQ