sjsreehari/Wi-Fi-gesture-dectection

WiFi CSI-based gesture recognition using dual-path ensemble deep learning (CNN2D + CNN1D-LSTM). 90.19% accuracy on ESP32 hardware with 426K parameters.

31
/ 100
Emerging

This system helps you detect human presence and specific gestures in a room using standard Wi-Fi signals, without needing cameras or wearable devices. It takes raw Wi-Fi Channel State Information (CSI) data from an ESP32 device and outputs classifications like 'Not Occupied', 'Occupied Static', or 'Occupied Motion'. This is ideal for smart home enthusiasts, facility managers, or privacy-conscious individuals looking for contact-free sensing solutions.

Use this if you need to detect human presence and gestures discreetly and without visual data, leveraging existing Wi-Fi infrastructure.

Not ideal if you need to identify individuals, perform very fine-grained gesture recognition, or require extreme precision in complex environments.

human-sensing smart-home-automation occupancy-detection contactless-interaction privacy-preserving-tech
No Package No Dependents
Maintenance 6 / 25
Adoption 5 / 25
Maturity 13 / 25
Community 7 / 25

How are scores calculated?

Stars

10

Forks

1

Language

Python

License

MIT

Last pushed

Nov 10, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/sjsreehari/Wi-Fi-gesture-dectection"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.