euaziel

euaziel / WiFi-CSI-Human-Pose-Detection

Public

Human pose estimation using WiFi Channel State Information (CSI) and deep learning — enabling camera-free sensing through walls.

43
20
100% credibility
Found Mar 03, 2026 at 43 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

An open-source system that analyzes WiFi signals to detect human poses, vital signs like breathing and heartbeat, and presence through walls without cameras or wearables.

How It Works

1
🔍 Discover WiFi magic

You hear about a clever way to see people's movements and breathing through walls using just your home WiFi, without any cameras or gadgets to wear.

2
📥 Grab the free tool

Download the simple program that turns your WiFi into a smart watcher, ready in seconds with one easy command.

3
Connect helpers
🛒
Buy boosters

Get affordable mini WiFi sensors to place around rooms for seeing poses and heartbeats clearly.

🏠
Use home WiFi

Start right away with your regular WiFi for simple presence and motion alerts.

4
🚀 Watch it work

Launch the watcher and see live drawings of people's positions, breathing waves, and heart rates appearing on your screen.

5
📱 Check on loved ones

Set gentle alerts for falls or quiet moments, keeping an eye on family through walls safely.

❤️ Peace of mind

Rest easy knowing you can watch over elders or detect anyone in any room, all privately with WiFi waves.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 43 to 43 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is WiFi-CSI-Human-Pose-Detection?

This Rust project uses WiFi Channel State Information (CSI) for camera-free human pose estimation, detecting 3D human poses, breathing rates (6-30 BPM), heartbeats (40-120 BPM), and presence through walls up to 5m. Developers get real-time sensing via cheap ESP32 nodes or commodity WiFi, with outputs like 17-keypoint poses and vital signs streamed over REST APIs or WebSockets. One Docker command spins up a live demo at localhost:3000, no special hardware needed for RSSI-based presence.

Why is it gaining traction?

It stands out for privacy-first sensing—no cameras means no GDPR headaches—while delivering human pose classification and estimation that works in darkness or behind obstacles where vision fails. The Rust rewrite hits 54K fps processing with a 132MB Docker image, plus self-learning models that adapt across rooms without retraining. Devs love the crates.io packages for easy integration and one-liner verification scripts proving the pipeline.

Who should use this?

IoT engineers building smart home occupancy or elderly fall detection without wearables. Robotics teams needing through-wall human detection for cobot safety zones. Disaster response developers prototyping camera-free survivor localization with vital signs triage.

Verdict

Promising prototype for camera-free human pose estimation, but 43 stars and 1.0% credibility score signal early-stage maturity—docs are thorough with ADRs, 542+ tests pass, yet real-world CSI hardware validation is key. Try the Docker demo if wireless sensing intrigues you; skip for production until more benchmarks surface.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.